Trigger warning for this. For, like, all of this. Just, it's just, don't even read it. Go home.
I. What's Okay?
Many years ago I took part in a discussion that would fundamentally redefine the way that I looked at fiction from that point on.
The topic of the discussion was the inclusion of rape in tabletop games.
The original poster made the argument that, while rape is horrific, so are torture, murder, arson, etc - all widely accepted as fare for tabletop gaming campaigns, even lighthearted ones. Attempting to draw a line about the inclusion of rape suggested that it was implicitly, intangibly more heinous than torture, murder, etc. There were essentially two counter-arguments being made: "rape IS more heinous than torture, murder, etc., which is why those are acceptable and rape is not", and "they're the same, in that they're all unacceptable".
We can essentially chart this argument by creating three points on a line.
The LEFT SIDE is "all acceptable". The premise of this viewpoint is that "it's just fiction" is a common, accepted idea. It is a view that can be described as distasteful, but not inconsistent. It is an idea that fiction lies outside our "real morals", and thus real morals cannot be applied to what is purely fantasy.
The MIDDLE SPECTRUM is the idea that some things are innately okay, and some things innately are not. Since it is a spectrum, this covers a wide combination of "some things are okay, others aren't". This view is best described in terms of its objective weakness: its principles are established based on personal feelings, rather than a legitimate guiding principle. However, due to the nature of our society, it is by far the most common viewpoint.
The RIGHT SIDE is "all unacceptable". While the Left Side operates under the argument that "if it's fiction, it's okay", the Right Side uses the concept that fiction does matter, and does have moral bearing. Therefore, if you wouldn't do it in real life, you probably shouldn't glorify it in fiction.
The original poster's goal was to convince people in the Middle Spectrum that they were hypocrites, while also expecting their enjoyment of murder & torture to override their disgust at rape. The Middle Spectrum individuals would then be shunted over to the Left Side, so they could continue enjoying the distasteful things they liked without feeling hypocritical about it. However, in the case of some individuals (myself included), the disgust for rape overrode the enjoyment of murder & torture. As a result, we expanded our feelings of guilt and disgust to include acts of murder and torture. This pushed us away, onto the Right Side.
This basic argument is the foundation of all "it's just fiction" arguments that have ever, or will ever, happen. You either accept it, you don't, or you muddle in between picking and choosing.
II. Criticism of Sexism vs Criticism of Violence & Issues of Severity
Beginning primarily in the early 90s, videogaming was constantly under fire for the depiction of violent, gory, or otherwise distasteful content. The most common assertion - or at least the most prominent assertion - was that violent games would transform a regular human being into a murderous psychopath. This idea is commonly confronted by gamers specifically because it is easy to disprove; there are numerous studies that debunk the idea that violence in games leads directly to violence in real life.
When critics of sexism in games bring up concerns that the depictions of women and sexuality in games will propagate certain ideas in the gaming public, the counter is often that video games are "proven" not to affect people. Which is to say: "if violent games don't affect people, how can sexist games?"
I've seen quite a few critics have a difficult time with this question, and they usually have a difficult time because they're trying to make excuses for violence in games. The most common explanation is that "murder" is an act, whereas "sexism" is an idea - it's easier to propagate sexism than it is to commit murder, and it's easier to convince someone to behave in a sexist way than to convince someone to commit murder.
The argument is fundamentally sound, but in the context of the discussion, it's simplistic. It does this because it is trying to encourage one discussion (sexism in games) while quashing another (violence in games).
MURDER is the most extreme actualization of the concept of HATE.
RAPE is the most extreme actualization of the concept of SEXISM.
Saying "games don't cause murder" is like saying "games don't cause rape": it's provably true, but there are many ways for a concept to affect people without pushing them to the furthest possible reaches. Refusing to talk about violence in video games except in terms of "murdering people" is like refusing to talk about racism in media except in terms of "lynching people". There's a lot of room between "absolutely no change" and "the most extreme change possible". There are plenty of studies that indicate violent video games can increase aggression, and it's common sense that cultural depictions affect people's perceptions of the society around them.
It's also worth noting that murder and rape tend to fall under the same primal concept: the desire for power. People enjoy killing in video games because it feels good to be better and stronger than other people. A similar motivation exists for rape in fiction; it's certainly not about the sex, because the sex itself is fabricated, and could be totally consensual just as easily. But it's not, because that's not what's important. Rape is about power, just like killing is about power.
III. Comparing the "MTR" Triad
These are the three most prominent "immoral acts" in games: Murder, Torture, and Rape. They are written in order of ascending vileness; murder is the least bad, torture is more bad, and rape is the worst.
Here's an example sentence regarding morality in fiction:
"Yeah, I know it's bad in real life, but in fiction it's okay."
Going back to this article's Point I ("What's Okay?"), one of the defenses I've heard from the Middle Spectrum is the idea that murder and torture are widely accepted as "bad", whereas rape is still a common issue and thus more dangerous with regards to influencing people in real life.
However, I don't agree with this idea. In fact, the specific order of the MTR triad reflects how commonly accepted actions are in real life (and, accordingly, in fiction).
MURDER in real life is easily excused by a huge number of scenarios, many of which even strip the act of the name "murder". If you kill an enemy, that's not only "okay", it's encouraged. If you kill an attacker, that's okay. If you kill a criminal of pretty much any sort, that's considered okay - and this one forms the slippery slope, because you'll see people encourage the shooting of protesters and other perfectly legal inviduals under the ASSUMPTION that they're doing something illegal. The United States of America in particular has a massive legislative bloc built around the idea that private citizens have the right to bear arms - which is to say, private citizens need to be able to commit "justified murder", because there are so many scenarios in which that need might arise. Real pacifists are few and far between, and they're massively outnumbered by people who think that killing is an acceptable choice in a pretty wide number of scenarios. It is therefore extremely simplistic to say that people generally accept that murder is "wrong", and more accurate to say that it is commonly glamorized, glorified and anticipated.
What does that mean for this comparison? Killing in games is not only common, but more often than not it is depicted in a purposefully unrealistic manner ("shoot bad man, bad man fall down"). Killing, as an action, is not "shocking" in games, or "jarring", or "upsetting", unless a game is specifically going out of its way to create that effect. And as games have gotten more realistic, we begin to associate more in-depth depictions of murder with simplistic black-and-white morality - Sniper Elite being the most prominent example of that. Despite having incredibly in-depth models of the human body being penetrated and torn by gunfire, the game operates on the same basic moral assumption that fueled Wolfenstein 3d: "it's okay to enjoy killing if you're killing bad guys". So you end up with cases where even brutal, visceral murder is associated with clean, justified morality.
TORTURE in real life is generally discouraged by society, but exceptions always exist. The Jack Bauer concept of a "ticking time bomb" convinced many people that torture was (a) effective and (b) necessary, and if we took away the CIA's right to commit torture, we would end up in a scenario where we could not effectively protect ourselves. This argument was so persuasive that it was cited by Antonin Scalia of the United States Supreme Court - yes, it was a fictional scenario in a fictional show, but many people were convinced that it was hypothetically plausible. Of course, the reality of torture, especially the CIA's torture, is that it is (a) generally ineffective and (b) pointlessly, needlessly, grotesquely cruel. The "ticking time bomb" scenario almost never arises and is thus statistically irrelevant to actual torture issues.
When torture shows up in games, it is almost always of the "brutal interrogation" variety. From "Splinter Cell" to "The Punisher", the idea of horrifically hurting a flat, one-dimensional "bad guy" character draws into the same wellspring of hatred that fuels the act of killing, and the act is morally justified because prisoners provide intelligence. It's not needlessly cruel, goes the implication, but an act of genuine tactical necessity. Therefore, games feel okay showing torture as an extension of their existing paradigms; you kill because you have to, you torture because you have to. It's all necessary for self-preservation and/or saving the world. Also, they're bad guys - they deserve it anyways. They'd undoubtedly kill you if you left them alone. So what's the harm? By presenting unlikely situations and totally dehumanized enemies, fiction is capable of distorting the public's view about real torture and what it entails.
RAPE in real life is "almost never okay". Without getting into the network of fringe philosophies, it's generally accepted, in our society, that "rape" is a universally bad thing. If a soldier kills an enemy, that's justified; if a soldier tortures an enemy, that might be justified; if a soldier rapes an enemy, that's weird. Games commonly feature killing as a mechanic, and sometimes feature torture as a mechanic, but the inclusion of rape in a game generally only exists if the game is rape pornography.
There are a lot of nuances of the depiction of rape that we could talk about. For example: most rapes in fiction involve strangers ("thugs") when in reality this is a small percentage of the total. The reason most rapes are depicted as "thugs in alleys" is because there are very clear ways to deal with "thugs in alleys" that coincide with conservative values ("don't dress a certain way", "carry a gun", "don't do drugs", "don't be a prostitute"). As a result, the "thugs in alleys" model is used as a coercive threat - "women, do what we say or else rape will happen to you". Often, conservatives will dismiss other types of rape because they suggest an alternate problem - a cultural problem, a patriarchal problem, a communication problem, etc etc etc. "Rape" is only useful to conservatives when it is "thugs in alleys", which is one of two reasons why "rape in alleys" is so common in fiction.
The other reason is because "rape in alleys" is an easy conundrum easily solved by violence, which is a thing that fiction loves. Superheroes solve alley rapes because that is the only thing they are good at. We don't have superhero comics about systemic reform and revitalization efforts; we don't have superhero comics about legislation and education. We have superhero comics because (a) we want to enjoy violence and (b) we want to justify this violence as good and necessary and heroic. "Rape in alleys" fulfills that condition by being easy and simple in a way that most real rapes aren't.
I could also talk for quite a while about the "no means yes" angle of certain rape fantasies and why that negatively impacts a "positive consent" culture. I could also talk about the "rape victim starts becoming aroused" angle, which is an incredibly volatile issue in real life and is often used to justify an act as "not being really rape". But, believe it or not, I'm actually getting off-topic.
The point of the MTR comparison is this: rape is unacceptable in most games, torture is sometimes acceptable, and killing is almost always acceptable. Very relatedly, rape is almost always unacceptable in real life, torture is sometimes considered acceptable, and there are a wide variety of justifications for murders. The representation of "bad things" in fiction matches up pretty well to the justification of those "bad things" in real life.
Games do not commonly feature killing by accident, they do this because society, in general, accepts the idea that it is Okay To Kill Bad Guys. The fact that it's okay to kill bad guys means that depictions of killing (whether simple or detailed) are common in our media. Eventually, games move on to the idea of killing people who AREN'T bad, and justifies it because it's "just fiction". The missing piece of that transition is that people are already okay with the idea of killing people. It's not shocking or disgusting because we've already been exposed to the idea in a safe, justified environment.
However, rape is NOT commonly depicted in games because it can't be justified. Therefore, unless you're into rape pornography (and a lot of people are, especially in the gaming world), the imagery of rape is probably going to be jarring and disgusting to you, because the act of rape itself is jarring and disgusting. By default, a human being watching a rape is most likely going to think it's horrific, in the same way that by default a human being watching a murder is most likely going to think it's horrific. For killing, repeated exposure to "sanitized" fictional murder has created a smoother, less jarring experience, but that process hasn't happened for rape. So the people who are into rape pornography, who have already been desensitized to the act, are going to be like "I don't see what the big deal is", while everyone else has horrific visceral reactions to one of the most objectively awful things that can happen to a human being.
That is the difference between rape and violence in games.