Everything You Always Wanted to Know About Sex in Video Games (But Were Afraid to Ask)

Part of a series on The Caldecott Caper. 

Okay! So, this is a topic I’ve been thinking about for a long time now, and as I’m starting to write this post I’m trying to decide how much I want to stick to describing what I did in Caldecott, and how much I want to discuss my reasons for doing it. I'll probably end up muddling through both of them.




I’ve been a big fan of video-game romances ever since Baldur’s Gate 2. Maybe even before that - part of the reason Final Fantasy VII was so memorable to me was taking Aerith on a date at the Gold Saucer. In much the same way that the SNES Shadowrun game blew my mind by introducing the idea of setting a roleplaying game in something like our own world, BG2 blew my mind by introducing the idea that video games could portray realistic human relationships.

Like a lot of BioWare fans, I’ve avidly followed their increasingly confident handling of romance against a changing cultural backdrop. They’ve always been trailblazers, from offering complete romantic arcs in BG2 through their  depiction of same-sex relationships and fully-realized transgender characters. As much fun as the games have been to play, I’ve almost been more fascinated by the behind-the-scenes thought that they’ve put into these relationships. Their fans have HUGE emotional investments in these parts of the game, and BioWare has been particularly thoughtful and deliberate in how they honor the trust they’ve been given.

One major turning point for me was watching David Gaider’s seminal GDC talk on Sex in Video Games. If you have even the slightest interest in the topic, I highly recommend watching it yourself. David is funny and humble and incredibly perceptive, telling an insider’s story about all of BioWare’s attempts, failures, and successes over the years.

I knew that I wanted to write a romance even before I knew for sure that I was going to write Caldecott. Nearly a year ago, I was tossing around a few different ideas - doing another Shadowrun campaign, or teaching myself the Divinity: Original Sin editor, or taking the plunge and learning Unity. My number-one objective, though, was to try my hand at writing a romance. I ultimately decided to stick with Shadowrun in no small part because I could visualize how I could make that romance “work” within the tools I had already learned, rather than starting over from scratch.

I’m definitely standing on the shoulders of giants here - I have the benefit of having seen the failures (whether artistic, moral, or financial) made by others, which gave me a clearer picture of how to try to avoid repeating those mistakes. At the same time, I’ve developed some strong opinions and theories over my years as a fan of romance, and welcomed the opportunity to put some of them to the test.

So, let’s start by talking about gating! Back in the benighted days of BG2, your character needed to match a specific set of requirements to even be considered by a potential partner: to “date” Aerie, you needed to be male, belong to a particular race (human, elven, half-elven or gnome), and be of neutral or good alignment. Interestingly enough, one of the very first player-created mods to come out for this game was the “Can’t Kill Love” mod, which removed all such restrictions. This had some occasionally-hilarious side-effects; most notably, Viconia’s man-hating dialogue makes no sense when delivered to a female lover. But I think it also spoke to the desire for players to pursue the objects of their affections, even when that didn’t match the original vision of the game’s designers.

Over the years, BioWare moved gradually in the direction of more flexible love interests. In the first Mass Effect game, Liara is a member of a mono-gendered alien species, but absolutely reads as female; the game was famously reviled on Fox News for including cut-scenes showing a female player character kissing her. Interestingly, there was much less controversy two years later in Dragon Age: Origins, where the male elf Zevran would make love to both male and female players.

This design reached its apotheosis in Dragon Age 2, which had four romanceable characters (not counting DLC), all of whom were bisexual. There was a TON of controversy around this, which I don’t want to get into here. Many players appreciated finally being able to “romance all the things”; others complained about how unrealistic the ratios were; others were uncomfortable with the way the game downplayed the characters’ bisexuality, only focusing on their attraction to the player’s gender.

Since then, BioWare has continued to grow its stable of romanceable characters, and along the way has grown the constellation of sexual orientations on display. The choices used to be “heterosexual or bisexual”; in the latest Mass Effect and Dragon Age games, there are also homosexual characters who can exclusively be romanced by players of their own genders.

To be honest, I haven’t decided for myself which I prefer: the “everyone can love everyone” free-for-all of DA2 or the “everyone is represented” mosaic of DA:I. When it came to Caldecott, though, I felt slightly hemmed in by practical constraints: I’m just a single writer/developer, making this whole huge thing all by myself in my spare time. Writing a full six romance tracks just wasn’t feasible, for both scheduling and plot-related reasons.

As much as I knew that I wanted to write a romance, I also wasn’t fully confident that I could do so successfully. I’ve never written a romance before, in any medium - not in prose or fan fiction or text adventures or anything. So, from the very beginning I was careful to write in such a way that, if I was unhappy with the final result, I could completely remove it and still have a full, functional game. I was more or less fine with taking this risk with two romance tracks, but just couldn’t justify the effort involved in writing additional tracks that might be scrapped, even apart from my worry of romance fatigue.

Fortunately, the two characters I had in mind for romance were both very believable as bisexual characters. (Spoilers incoming if you care about that.) Sable’s defining characteristic was his curiosity: he’s constantly searching for new experiences, trying to discover first-hand everything that the universe has to offer. It would be strange if he WASN’T bisexual. Persi would make sense as a lesbian or as a hetero woman, but, well, she isn’t either. I mentally put her at about a 4 on the Kinsey scale - she’s had some bad experiences with men in the past that leads her to be more comfortable around other women, but she’s attracted to people based on their individual characteristics and won’t turn down the right guy.

One funny little epiphany that I had while writing the romances was how relatively easy it was to write gender-neutral romance content. It would have been much simpler to only permit human romances but allow both men and women than it would have been to permit only male romances but permit dwarves, humans, elves, orks and trolls. Of course, I was allowing both variables to change, and regularly found myself wondering, “Does this make sense? What if the player is a dwarf? Can he or she reach that high?”

It’s kind of a truism that challenges lead to creativity, and I definitely found that to be true here. Something that sounds incredibly simple, like “You kiss her,” becomes really complex when the player can be pretty much everything. What if you are 3 feet tall and she’s 6 feet? Are you standing on a ladder? Is she crouching down in front of you? Does she pick you up to smooch her face? And none of that makes sense if you’re a troll: now you’re leaning down to her, not up to her. I didn’t want to write three versions of every dialogue, so I found creative ways to collapse the problem. Suddenly there’s tickling involved, and you’re both on the floor, and the scene is simultaneously much easier to write and more interesting than it would have been originally.

One recurring theme throughout development was a fresh appreciation for everything BioWare does. They’ve mentioned before how challenging it is to support romances for the different fantasy races of Dragon Age, and, boy, can I believe it. I don’t even need to animate video or anything, and already run into enough problems when I’m doing nothing but writing text.

Along the same lines - romance was one area where Shadowrun’s text-heavy approach really shines. Even high-powered AAA game engines can have a really hard time depicting seemingly simple interactions like touching or kissing. Since players are more or less buying into the text, though, I could get away with writing some scenes that would be very expensive to animate.

At the same time, I tried to use all of the tools at my disposal. Shadowrun has a simple but pretty powerful animation system; it’s under-utilized in the official campaigns, so it carries a big impact when a character starts dancing, or lies down, or otherwise interacts with their environment. I could also draw on standard visual storytelling techniques like fading out, using dramatic lighting, etc. These took a little while to script, but they help the romances feel more special, and hopefully help the character connect more to their relationship arc.

Oh, yeah, arc! That’s another thing I wanted to write about.

As I said before, I’m very indebted to BioWare in general and David Gaider in particular for generously sharing their experiences with creating romances. David used to run a tumblr, in which he shared surprisingly detailed notes about how his team constructed romances, down to a very technical level of showing how they traced out the arcs of the characters and tied them to developments in the plot. Reading that was eye-opening to me.

I didn’t exactly copy the system he described, in large part because of the difference in how Dragon Age and Shadowrun games are structured. DA tends to be more open-ended, with the player free to go pretty much anywhere and pursue any one of many quests at any given time; Shadowrun is more of a chain of missions, with some flexibility in how they are put together but without the ability to laterally switch between quests mid-stream. Still, his posts got me thinking about what was actually going on mechanically in all of those romances I’ve enjoyed, and ponder ways to translate that impact to Shadowrun.

The standard progress of a romance is a series of answers to questions: “Do you like me? Do you LIKE-like me? Do you like me more than anyone else?” There’s a gradual escalation in intimacy and affection, eventually putting your character on the path to commitment. It helps ensure that the player is buying into the romance, feeling like they’re making a choice and not being railroaded into something just because the game designer likes it.

I’ve thought for a while that this is actually a rather canny bit of psychology. There’s a potential for cognitive dissonance here: if the game gives you constant choice, and you end up in a romance then it must mean that you like the romance, right? Otherwise you’re making bad decisions, and you’re not the sort of person who makes bad decisions.

It’s also clever because it does a great job at emulating how love works in real life. Love isn’t a sudden bolt from the blue that strikes you and changes your life: love is a series of affirmations, a series of “Yes”’s given over time that gradually binds people together. In a video game, we approach this relationship through a keyboard and a mouse, but I think the principle is more similar than we might initially think. Love is something we practice, it’s something we work at, something we earn rather than are given.

Anyways. By making the player say “Yes” over and over again, we make sure they’re on board the love train before it departs the station. If not, that’s okay; this is an RPG, not a romance simulator, and they’ll still have fun. If they’re on the train, though, then they’re committed, and there’s a whole host of benefits that the player and designer get: greater emotional investment, a heightened sense of  stakes, and more emergent storytelling as your partner takes to the battlefield.

Okay! That’s more than enough theory. Let’s move on to mechanics.

First, gating. Any time that anyone suggests that Harebrained Schemes should maybe add a romance to one of their games, that suggestion is immediately dogpiled by lots of players angrily talking about how romance has no business in Shadowrun. Which always makes me sad - I like Shadowrun, and romance, and would like to see more of both - but also made me think more carefully about how I wanted to present romance in my mod. I ended up borrowing an idea from Fallout: New Vegas. That game brilliantly used the Perks system to enable romantic options: if you were playing as a woman, you could take the Black Widow perk to unlock flirting options when speaking to male characters, or Cherchez la Femme to do the same for women. That seemed like a neat way to let players engage in romantic content if they wanted it, while completely avoiding it if they did not.

In my case, I handled this through an early dialogue with Rafik, your first companion. As part of a general “getting to know you” conversation, he will gently give you the opportunity to state a gender preference; you can indicate which sex(es) you prefer, or make it clear that you don’t care for love. Future flirting options are then optionally disabled based on your choice. I liked this because it felt more like a roleplaying than a mechanical choice: you aren’t selecting a sexual preference from a drop-down box, but revealing it in a frank conversation with a buddy of yours.

One question I pondered for a while was whether to restrict flirting options to potential love interests, or allow flirting with a larger number of people. I think the main benefit of limiting it is to more clearly indicate to the player who is a valid LI; I was a bit concerned that, if they saw additional options, they might not pursue any romance at all because they were holding out for another one that they preferred. I experienced some (very mild) frustration in DA2 when I realized that I couldn’t romance Aveline, and in DA:I at being unable to romance Scout Harding, despite flirt options being available for both characters.

The counter-argument is that it’s more realistic to allow flirting with more people, and allows a greater range of role-playing. If you feel like your character would really like a particular person, then you might be annoyed that you don’t have the opportunity to express that affection. And even if they aren’t a valid LI, the other person could still turn you down, as so often happens in the real world. (Which then leads into a counter-counter argument, which is whether it’s really fun to let players experience romantic rejection in yet another medium.)

I ultimately decided to restrict flirt options to the two romanceable characters, but will be monitoring the reactions this decision gets… I doubt I’ll add more flirts in, but will keep those reactions in mind if I ever decide to try my hand at romance again.

In the Dragon Age and Mass Effect games, flirting tends to be a single-choice option: if you like the person, you hit the button with a heart on it and proceed. I do like the clear iconography of showing which options will encourage romance; it was annoying in Dragon Age: Origins when Alistair would mis-interpret my friendliness for affection. But I wanted to have more choice in my romances, so every time you have the option of flirting, you get not one but two flirt prompts to pick from. This gives players more flexibility in how they role-play this aspect of the story: are they a witty charmer? A confident admirer? A sweet partner? This also gave me more flexibility in the prompts I wrote; I could give players the option of using a really bad pun, without demanding that they use it to advance in their relationship.

The two flirt options are equally “good”; they each lead to different dialogue, but the main thing they’re doing is registering your character’s interest in the other character. After you have flirted at least twice and you’ve progressed far enough in the story, you will get the option to start your first “lovetalk”. I borrowed/stole this term from Baldur’s Gate 2, where this variable is used to track your progress through the four romance plotlines.

Persi and Sable have very different personalities and the content of their romances is entirely different, but structurally they are identical. After your two or more required flirts, your first lovetalk will broach the possibility of a more serious relationship. This offers some easy outs to players who mis-clicked or have changed their mind: were they just playing around? Are they willing to have a go at this?

This is also where I collapse the romance choices from two down to one. A small but persistent criticism of BioWare romances is the lack of polyamorous options. Patrick Weekes and other writers have frankly stated that this mostly comes down to scripting: allowing multiple simultaneous romances leads to exponential complexity that becomes impossible to write and even more impossible to debug. With just two options, I had a better chance of making it work, but was reluctant to take that risk for my first outing. So, that first lovetalk will also check to see whether you’ve started flirting with the other love interest, and if so, ask you to make a choice.

Narratively, this works pretty well. Persi is highly individualist - she barely feels comfortable letting one other person into her life, let alone two more. Sable makes it clear that he’s down for whatever group activities you might have planned, but that he won’t force Persi into anything she’s uncomfortable with. So, while I do allow players to express their preferences, they’ll ultimately need to pick between one of the two tracks (or jump off the train altogether).

There are a couple of throttling mechanisms in place to keep players from running through the romances too quickly. It would be weird if, say, you recruited Persi, then talked to her a couple of times, and then immediately started a relationship. (Okay, maybe not that weird in the context of Shadowrun, but weird for the arc I had in mind.) You’re always limited to a maximum of 2 substantive conversations with each companion during each hub visit, and can only start certain conversations after completing a particular number of missions. The upshot of this is that, if you recruit Persi first, you can start flirting with her immediately and begin the love track after moving to Halferville; but if you recruit her last, you’ll be flirting with her at the midpoint of the game, and only start “dating” her later. I did a lot of careful work with the throttles here, trying to help things have a somewhat natural flow while also accommodating the flexibility I wanted in the game  design.

After that initial gating conversation, Persi and Sable each have 3 additional lovetalks; as noted above, the content is completely different for each, but they follow an identical dramatic arc.

Lovetalk 2 is rising action. You’ve affirmed your affection for one another, and start exploring the benefits of your relationship. There’s more physical contact, more intimate conversation that starts to get at your personal feelings in a way that you wouldn’t explore with mere friends. It ends on a high note, cementing your mutual attraction.

Lovetalk 3 is the complication. Anxiety and conflict comes bubbling up, as your partner gets cold feet. This is centered on a fear of change: the increased intimacy is starting to change them in ways that worry them. This is a last chance for players to break off the romance if they aren’t happy with how it’s proceeding. Otherwise, though, they can talk through the problems with their partner, reaffirm their commitment, and create a shared vision for what this relationship could look like.

Lovetalk 4 is the climax. I was inspired by the capstone quest design for romances in Dragon Age: Inquisition, where you are set on a mini-quest that reflects your partner’s personality and desires. We are playing a video game, after all, and it’s good to use those tools to support the romance, rather than just having it be a sort of vestigial appendage to the gameplay. This is also yet another case of the player affirming the relationship: not just clicking “Yes” in a dialog box, but actually taking the effort and thought to complete a meaningful task. At the peak of this comes the sex scene, and the romance is “locked in” (if not yet complete).

(The sex scene is, hopefully, tastefully done. It amused me to write this in a gender-neutral manner yet again; I’m curious how many players will pick up on that. It’s quite brief in comparison to the overall span of the romance, and miniscule in the context of the entire campaign, but hopefully evocative enough to resonate with players.)

And, from there… well, without getting too deeply into spoilers, it ultimately determines the end of the game. Let’s just say that once again I borrowed heavily from Baldur’s Gate 2, with your partner playing a significant part in your personal story and ultimately, potentially, featuring in your epilogue.

Along with the great guidance from BioWare, I also benefited enormously from Feminist Frequency's Tropes vs. Women in Video Games series, which has spent several episodes specifically on the topic of sex and romance in games. Originally I mainly focused on ensuring equal representation: if both men and women are potential LIs, have an equal amount of content, and are portrayed equally, then there’s no problem, right? But her more recent videos made me think more about the role that the romance played within the game; I feel like I was mostly on the right track anyways, but in some cases I decided to make small adjustments, and other times felt reaffirmed in tentative decisions I’d made earlier.

A few particular items that may be noteworthy:

• There aren’t any mechanical benefits attached to the romance. You don’t earn extra points or level up by bedding another character. It’s another aspect to your story, not to your gameplay.
• Sex isn’t a reward. Your partner won’t fall in love with you because you defeated the arch-enemy or gathered 1000 coins. Progressing in the romance requires attention to and empathy for your partner; to succeed in the romance, you must focus on the romance.
• Your love interest kicks ass. They’re an equal partner on your team, bringing their own unique mix of skills to the challenge, and have value as people outside of their potential appeal as a girlfriend or boyfriend.

As noted above, restrictions lead to creativity, and I found that to be true here as well. When I realize that I'm falling into a trope, I hope to recognize it, and then rewrite to avoid the trope. That's always a good thing! At an absolute minimum, it means that I'm not repeating the same storytelling techniques that have turned this into a trope, and will come up with something relatively new that may surprise my players.

On a broader note, I thought a lot about agency, throughout the development of the game but particularly when writing the romances. It’s a tricky balance to strike. Players want to feel like their actions have significance, like they’re a force in the virtual world, that their decisions have consequences. It can be very frustrating to express a desire and have that desire thwarted. And yet, in the real world, we have limited agency; thus, it’s much more realistic for the player to experience periodic resistance and conflict.

In Caldecott, I tried to handle this by setting each major character on their own separate wavelength. They express their own opinions and desires, which may not match the player’s own. In a successful relationship, you essentially harmonize with them: finding a mutually-agreeable path that lets you maintain your independent identities while simultaneously joining forces and creating something new together. That promotes the idea that the other characters are individuals not under the domination of the player’s whims, while also allowing the player to steer the story in the direction they wish to go.

Okay. This post is way too long and I’m officially rambling now.

Bottom line: This was really hard, but also incredibly rewarding. Romance is one of the biggest minefields a video game writer can enter, and I have no doubt that I’ll get a lot of negative feedback for it. (I’ve found it somewhat helpful to think of this as a form of penance for the innumerable times I have complained about a lack of options in BioWare games.) I have no regrets about doing it, though. After playing through it myself and seeing how the romance integrates with the player’s personal story, I feel validated in my initial theory about romance being an effective technique for raising the stakes and making the player more invested in the story.

I have no idea yet if I’ll try this again; a lot will probably depend on what kind of reception this outing gets. If I do try my hand at romance in the future, I would love to work with a larger stable of characters, ideally with a wider range of sexualities represented. That might also be a good point to finally start actually collaborating with other writers. Precisely because these arcs are so personal, they’re emotionally exhausting to write. I think I could benefit from sharing the load, as well as gaining in perspectives from writers from different backgrounds than myself.