Great characters are usually at their best when they're surprising us. A villain who is purely evil is never as interesting as a villain who sometimes expresses kindness. A killer with a code; a monster with a soft spot for kittens. Yet some video games this past generation just didn't seem to get that—they pushed us to play purely evil or purely good.

[Heads up: This article contains spoilers for the endings of Mass Effect, Mass Effect 3, BioShock, GTA IV and Return of the Jedi.]

Video game morality systems became very popular over the last gaming generation, and with good reason. They're an often fun way to turn the non-game parts of a game—dialogue, and the other bits in between combat—into more of a game. Suddenly, the way you treat that shopkeeper has some consequence. You were a jerk to him! That might come back to bite you.

While some games handled morality in interesting ways, others could often be awfully hamfisted about it. In retrospect, the hammiest fist of all belongs to the binary morality systems exemplified by Mass Effect's paragon/renegade system.

Imagine: At the end of Return of the Jedi, Darth Vader decides to surprise us all. He stands up behind the Emperor, ready to kill him and save his son Luke Skywalker. Oh, but nope, sorry! He actually can't do that. After a lifetime of evil, Mr. Vader doesn't have the paragon points necessary to do such a thing. He's been evil so long that he can only let his son die, then do whiskey shots with Palpatine while kicking the smoldering corpse.


That's how the ending of Jedi would've played out in the Mass Effect universe. In that series, players would score paragon or renegade points by acting virtuously or villainously. The more points they had, the higher their ranking would be. That, it seemed, was all well and good, and a fun way to track your character growth and assign him or her an identity. "I like my paragon infiltrator, but not as much as my renegade vanguard."

But for the majority of its run (though not, notably, in its grand finale) Mass Effect locked off some high-level paragon and renegade dialogue options unless you'd gotten your character to a certain level of goodness or badness. For example, at the end of the first game, a character with a high enough paragon score can talk the villain Saren into killing himself, negating one of the final boss battles. That kind of thing occurred throughout—I was never more frustrated than when my renegade character would find a blocked-off red dialogue option, telling me that Blade Shepard—who was a badass, reporter-punching renegade!—still wasn't badass enough to pursue this particular path.

Each time you made a decision that was "out of character," you'd be costing yourself points in your chosen moral specialization and potentially locking yourself out of future conversation options.

What was no doubt intended as a way to reward focused players became, over the run of the trilogy, a restrictive set of moral shackles that trained players to role-play a single way. The morality system was telling you that in order to see everything and play the game "right," you would have to role-play your Mass Effect character without any nuance. Each time you made a decision that was out of character, you'd be costing yourself points in your chosen moral specialization and potentially locking yourself out of future content.


Mass Effect's morality problem runs through the core of the entire game. Just last week, Stephen lauded the conflict with Wrex as "The toughest life or death choice in the gaming universe." And I totally agree with him; for most players, it was a really tough choice. It was a bold bit of storytelling, and easily one of the most memorable gaming moments of the last eight years. Yet the first time I played, thanks to BioWare's flawed morality system, it wasn't actually a choice. I had been playing down the moral middle, and as a result didn't have a high enough paragon or renegade score to talk Wrex down. He died, and I didn't even have a chance to save him. I know now that had I completed Wrex's sidequest, there would've been a third way to talk him down, but the fact remains that in Mass Effect, that kind of weird situation happens all the time; the conflict with Wrex is just a particularly memorable example.

BioShock was another offender in the morality wars of the last generation. The 2007 game made a big show of giving players choice, and not just any choice: The choice to save or kill a bunch of little girls. On the surface, it was a daring, provocative move. But thanks to the way the game was written and designed, it ran into a couple of problems.


First of all, as pointed out by game designer Clint Hocking in his now-canonical 2007 critique "Ludonarrative Dissonance in BioShock," there was a conflict between the harvest/save choice and the rest of the game's story. Hocking says that when he first started playing, he felt that the "bad" option—harvest the Little Sisters to gain more power!—was in line with Rapture's self-interested Objectivist philosophy. However, the initial thrust of the narrative—save Atlas' family! —wasn't. He could accept the lack of choice in that second matter as a limitation of the medium, Hocking said, but when the game revealed that the player was under the power of mind control and that even their illusory choice wasn't a choice at all, it wasn't fair play.

As Hocking put it:

Yet in the game's fiction on the other hand, I do not have that freedom to choose between helping Atlas or not. Under the ludic contract, if I accept to adopt an Objectivist approach, I can harvest Little Sisters. If I reject that approach, I can rescue them. Under the story, if I reject an Objectivist approach, I can help Atlas and oppose Ryan, and if I choose to adopt an Objectivist approach – well too bad… I can stop playing the game, but that's about it.

That's the dissonance I am talking about, and it is disturbing. Now, disturbing is one thing, but let's just accept for a moment that we forgive that. Let's imagine that we say 'well, it's a game, and the mechanics are great, so I will overlook the fact that the story is kind of forcing me to do something out of character…'. That's far from the end of the world. Many games impose a narrative on the player. But when it is revealed that the rationale for why the player helps Atlas is not a ludic constraint that we graciously accept in order to enjoy the game, but rather is a narrative one that is dictated to us, what was once disturbing becomes insulting. The game openly mocks us for having willingly suspended our disbelief in order to enjoy it.


BioShock's manipulation may not have bothered some players as much as it did Hocking, but the fact remains that the game didn't neatly fold itself around the idea of player agency. For the focus of this article, let's put aside Hocking's larger critique. Even if we only focus on the harvest/save choice, BioShock fell into the same kind of rigidity trap as Mass Effect, largely because it offered two very different endings based on the choices you made throughout the game. If you wanted your story to end with a "good" ending, you had to commit to saving the Little Sisters early on. It was possible to harvest them at first, then have a change of heart, perhaps after listening to Tenenbaum's reasoning and getting a fuller picture of what was going on in Rapture. But if you came to the realization too late, sorry, you were pure evil all along, and you wind up being the bringer of the nuclear apocalypse. If only you'd changed your mind a bit earlier!

For all the problems Mass Effect 3's ending may have had, I actually liked that it eschewed the binary rigidity of the morality system that had preceded it. For once, I had a choice unrestricted by the choices I had made leading up to it. My renegade character was free to choose the more self-sacrificing option, or vice-versa. While many felt that the ending betrayed the three games that had preceded it, I actually didn't mind this particular bit of divergence. The inflexible paragon/renegade morality system had, by then, become one of my least favorite aspects of the series, and if the ending was going to shrug off one of the series' traditions, I was okay with it being that one.

I'm picking on Mass Effect a lot here, but much of the last generation has been spent watching game developers wrestle with in-game morality. Fallout 3 did an interesting turn with the series' karma system, assigning positive and negative karma points to all sorts of in-game actions, from mass murder to petty larceny. It wasn't always the most nuanced system (maybe I'm stealing that water to give to a dying man on the wastes!) but it also wasn't all that consequential, when it came down to it. Fortunately, many of the game's big decisions didn't hinge on your karma rating—you were free to make them as you wished.


Sucker Punch's PS3 superhero game Infamous, and in particular its sequel Infamous 2, made some bold storytelling choices. By the end of the second game, Cole McGrath was either pure Electric Jesus or history's greatest monster. But while the wildly divergent storylines were confident and interesting, I always found it a bit annoying how the game tied high-level powers to Cole's morality.

That approach encouraged a second playthrough, but I'll never quite forgive the second game for tying the best locomotive power—the ability to launch yourself into the air with a blast of ice—to the "good" path. I'm still halfway through my evil playthrough and I just can't get that psyched to fly around because I don't have the ability to joyously leap tall buildings in a single bound.


Red Dead Redemption took a successful if somewhat less ambitious path by making its morality system apply to the game's open world but not to its story. If John Marston behaved like an outlaw, he'd gain notoriety and eventually attract deadly posses of lawmakers who would gun him down. But if he behaved as a saintly do-gooder, he'd be heralded as a hero all across the frontier. Neither approach had any affect on Marston's pre-authored story, which left players free to toy around with it. It definitely could cause the sorts of calm-cutscene-vs.-open-world-mayhem dissonance that Grand Theft Auto players have grown used to, but I enjoyed how RDR made the player's actions feel consequential without needing to rewrite its own story.

The cyberpunk RPG Deus Ex: Human Revolution tied various rewards and penalties to your in-game actions, but those interesting conversational boss fights—where you'd try to reason with or negotiate a hostile or guarded character into compliance—never changed based on your character's past choices. And Grand Theft Auto IV, as it happens, contained several of my favorite choices of the last gaming generation. Some of them had a dramatic impact on the story, while others were simply left for players to interpret. My favorite of the latter type came near the end of the game, when Niko was given the option to kill or spare Darco Brevic, a man who had gravely betrayed him in another life. Kill him or let him go: The choice had no practical consequence on Niko's story, but it was of immense consequence to Niko himself.


The examples go on and on, far too numerous to list here. And I'm leaving out the much more electric, human morality of online games like the zombie survival sim DayZ. That game put players into something of a real-life psychological experiment, often with fascinating results. Will any single-player game manage to capture the unpredictable human darkness of a DayZ? I'm doubtful. But it's been fun to watch them try.

A lot of video-game morality systems are drawn from pen-and-paper RPGs of old, where stats were assigned to all manner of conversational option, and the ability to charm and intimidate NPCs functioned as a reward for players who invested in their charisma or strength rating. There's an important difference between charm bonuses and narrative moral choices, if only in terms of believability. While a stat-boost in conversation is clumsy, I'll go with it to a point… but only to a point. I understand why my charismatic rogue might be able to talk down a shopkeeper's prices, but why wouldn't he be able to behave like a jerk anytime he wanted to? What do stats have to do with that?


I understand why my charismatic rogue might be able to talk down a shopkeeper's prices, but why wouldn't he be able to behave like a jerk anytime he wanted to?

While previous BioWare games—Knights of the Old Republic, Jade Empire—have also hewn to Mass Effect's "pick one morality and stick with it" way of doing things, this past generation the studio also crafted a notable exception: Dragon Age. I've always found the morality of Dragon Age's dark fantasy universe to be more compelling than that of Mass Effect, and I don't doubt that's in large part because it doesn't concern itself with giving your character "points" for goodness and badness. Some stats do allow your character to unlock unique conversation options, but they exist independently of one another—you can be charming and be intelligent, should you choose to be, and usage doesn't cause the stat to increase.


In Dragon Age, as in other successful choice-based games like Telltale's The Walking Dead and CDProjekt's The Witcher series, you'd be placed in a series of difficult moral quandaries and often must choose between the lesser of two evils to proceed.

Dragon Age: Origins did trip on itself in how it gamified party approval—your choices would often be met with approval by some and disapproval by others, but it was pretty easy to "gift" your way back into the good graces of even the most pissed-off follower. Fortunately, while other characters would sometimes make their own decisions based on your past actions, the game never really restricted your character based on his or her stats—no matter your past choices, you were given a full range of options.


Speaking of The Witcher, that role-playing series perhaps best exemplifies non-stat-based choices. In Witcher games, the only consequences are logical ones: You may kill a character, but if you do that, she won't be there to help you later. You may save a character, and he may tell you how to find a useful item. Or you may anger a character, who will then be unwilling to help you in the future. Your protagonist Geralt isn't given color-coded moral stats, and you'll never be unable to make a crucial choice simply because you didn't have a high enough ranking. If you want to choose to kill that guy, you can kill that guy. You've got a sword, and he's right there. Your call.

Here's hoping that in the generation to come, game developers continue to experiment with flexible worlds that react to players believably, giving us choices that run the gamut from red to blue to all the shades of grey in between.


Last-Gen Heroes is Kotaku's look back at the seventh generation of console gaming. In the weeks leading up to the launch of the PlayStation 4 and the Xbox One, we'll be celebrating the Heroes—and the Zeroes—of the last eight years of console video gaming. More details can be found here; follow along with the series here.

To contact the author of this post, write to or find him on Twitter @kirkhamilton.

Top image by limis, DeviantART