The lead writer behind BioWare's Dragon Age discusses how he crafts fantasy worlds
David Gaider has been building and defining fantasy worlds in video games for decades, most notably during his time at BioWare.
Currently at Baldur's Gate: Enhanced Edition studio Beamdog, we caught up with the writer as part of our How To Build A Universe feature, gaining more insight into how he helped bring worlds such as Thedas to life.
Read on to find out more about the importance of characters, the difficulty of presenting backstory and how medieval European history shaped the world of Dragon Age.
When you're creating a new fictional universe, how do you ensure it stands out? Anything inspired by high fantasy invites comparison to The Lord of the Rings, but how did you ensure Dragon Age wasn’t too Tolkien-esque?
Well, that’s a tricky question because I think that’s what we were going for with Dragon Age.
Part of it was working with artists because a lot of creating that distinctiveness fell to them. The ultimate goal is that when someone sees a screenshot of your game, they should be able to say ‘oh, that’s Dragon Age’ or ‘that’s Mass Effect’. I remember the Mass Effect artists were looking at particular hues and a certain colour range they could use with the world.
It's all about having certain distinctive parts that only exist in that world, without overwhelming it with so many distinctive things – then it becomes just a mish-mash to the eyes of a new player. If they walk in and everything is different, then they’re overwhelmed. But if a lot of it is familiar, and there are touchstones they’re familiar with, but then there are certain things that stand out as distinct, that’s going to be much more accessible to the player.
Where do you start when you’re trying to create a fictional world? What’s the first step?
A lot of time was spent going around the various team members and asking what they wanted to include. If you’re starting with a complete blank slate, you ask yourself what you want it to be. If the artists want to draw certain types of characters or certain looks, then I have to make sure I include those.
In this case, we started with somebody like James Ohlen, who was the lead designer on Dragon Age: Origins, and he wanted a certain type of fantasy that was reminiscent of Tolkien – so a pseudo-medieval Europe. I remember he gave me a book on medieval European history to go through and make sure there were points of history that were drawn into the world, so they would be familiar without replicating them entirely.
Once I had all that information collected, in terms of what the rest of the team wanted to see, it was a case of sitting down and drawing out an outline. I approached it the same way I’d start writing a game: starting with an outline, throwing about some names and so on. Ferelden’s actually one I came up with towards the start; I wanted something that would be like medieval Britain, but in the south. I wanted France to be the major nation culturally and militarily, and that became Orlais. I wanted a counterpart to the waning Roman Empire, which would eventually become Tevinter.
I may not have known what the story was going to be for the initial game, but once I had Ferelden, Orlais and the Tevinter Imperium in my head, I started to get a sense of Thedas as a real place.
I think the first thing I did once I had a simple point-form drawn up was I drew a simple map. Once I had a map, it made it easier. Although a large part of that is whenever I read fantasy books, there was always a map either at the front or back and that made it physically real to me.
Science fiction doesn’t have the need for pretty maps, but I imagine it’s still good to get that sort of information down before you start. I suspect the process for Mass Effect was fleshing out enough detail that the galaxy started to feel like an actual place, like a real setting.
I may not have known what the story was going to be for the initial game, but once I had Ferelden, Orlais and the Tevinter Imperium in my head, I started to get a sense of this as a real place – to the point if I sat down and wrote about it, I could have people within the game discussing the world and having opinions on various places.
It's surprising that you drew the map because you’d think that would be the artist’s duty. Rhianna Pratchett recently told us writers should be involved in all forms of games development – does that apply to universe creation?
For Dragon Age, I was left mostly to my own devices but I remember I did another world for BioWare’s new IP and that one was much more collaborative, which I actually liked more. Having a lot more input from the team, getting a lot more feedback on what they want, stopping at various points to see if the world worked for gameplay – I found that interesting.
Having the writers in at every level works both ways. As a writer, I don’t want to go off and work on something in a bubble so that I end up with a setting that makes for a really nice story but not for great gameplay or doesn’t fit in with the systems we’re planning on. One of the problems with Dragon Age was I didn’t put enough deserts and wastelands in there, and when we started work on the game the artists were keen on doing deserts for some reason. It would have been nice to have more of the team involved rather than just at the very beginning.
In the same way, having the writers be there when the rest of the team is working on other elements is also useful. If the systems designers are there and deciding what character classes or skills will be included, having a writer in the room will at the very least help them incorporate them into the history. We didn’t get that at the beginning of Dragon Age so you’ll notice that a number of the classes or specialisations did not actually make an appearance in the history or the world itself until much later in the series. It’s because they were done without any discussion with me.
So from both sides, it’s always good to discuss things and get everyone on the same page. I know a lot of companies bring in writers after the fact. It’s sometimes disappointing to see designers build a whole game and all its system then hand it to a writer and say, ‘hey, build a world around that’. I mean it can be done, and has been done, but you end up with something that feels very tacked on. Those are teams that expect writers to come in and with their magic writing wand make things better. But writing is always better when its done collaboratively.
How do you balance giving your world, your lore, your backstory and so on depth without bombarding players with details that ultimately don’t matter to them or what they’re immediately doing?
It gets tougher as a series goes on, because you want to both make it accessible to new players and avoid insulting the fans, who want be very patient if you take the same process as you do with the first game. If you repeat everything and ease players into the world every time – that’s what you need at the beginning of the series.
You don’t want to assault the player with names. We went through a lot of iterations at the very beginning of Dragon Age Inquisition to try and avoid having Cassandra bombarding you when she’s explaining what has happened: the Chantry, the Divine, the Tevinter – even the Grey Wardens came up at one point. There were a lot of places and groups mentioned, and even if you tried to use them in context, all the players will hear a load of names and not fully understand what’s going on. That gets tough, especially later on in the series where some kind of knowledge is assumed.
It’s sometimes disappointing to see designers build a whole game and all its system then hand it to a writer and say, ‘hey, build a world around that’. I mean it can be done, and has been done, but you end up with something that feels very tacked on.
But I think a lot of that is made easier by paying attention to the information flow. You have to almost drip-feed your lore, and try to avoid requiring the player to know these things. Just tell the player exactly what they need to know to complete their next task. They don’t need to know the background, the lore of the political mechanisms around it – that can be going on, and you can make investigation into it optional for the player. That’s the biggest challenge for a narrative designer, though: dealing with the information flow, especially at the start of the game.
You must come up with so much backstory and content, but how much never makes it to the final game?
I’d say about half. Part of it is that we always come up with more ideas for plots and stuff than actually work out. Some are cut before they’re even started, some are scrapped because they’re just not working, and then there are some where the plots themselves are cut down and need to be put later in the game because they have information flow issues.
The prologue for Inquisition was written entirely from scratch three times, if not four. Ignoring that, even when I look at the final version it’s about a third smaller than it once was. We had to cut the number of conversations being had, there was let of pre-amble about the political significance of the meeting that was destroyed – did the player really need to know all that?
There’s a struggle when you’re a writer to impart as much information as the player would get from reading a book – but the information isn’t absorbed in the same way. We tried to shift a lot of that information into things that are optional, so players can find it if they try to look for it.
Even then we had issues: putting things on question hubs actually led to a lot of players exhausting the hub because they thought that was what they had to do, even if they don’t like spending that much time in a conversation. In testing, we’d have people go through every single question in case there was something important to be learned and then complaining that the pacing was too slow. It was a bit catch-22 from a design standpoint: making things optional isn’t always the answer, so a lot of information was cut because it was unnecessary. Later on when you have the player interested, when they’re well into the game, maybe that’s the best time to start including a lot more optional information.
So how do you avoid 'world-building disease'? How do you stop yourself from writing too much content if half of it is going to be cut?
It’s not always a bad thing. If the developer or the writer has only created the part of the world that the player directly interacts with, that becomes readily apparent. When a writer knows way more than the player ever sees, there’s a sense that there is lot more happening in the world that just what they experience themselves. I think it’s important that this exists, so I don’t think it’s possible to develop too much.
How do you ensure all of your characters, not just the companions but the protagonist and antagonist, are memorable and avoid clichéd archetypes?
Do we do that? There are people that argue that BioWare keeps having the same type of characters – and to a degree, that’s true. But if you blur out the differences between characters, then of course they’ll fit into categories.
If you want the player to save the world, it’s hard to make them care about that because the world is this big, vague concept.
The point isn’t to try and make them not fit into an archetype, but to make them have relevance to the player. When we sit down and design a character, we’re not trying to work out how to make that character unique in every possible way. We’re trying to work out what the character is going to make the player care about and how they connect the player to part of the world or story. If the companion has no major connection to the world, why are they even there?
If you want the player to save the world, it’s hard to make them care about that because the world is this big, vague concept. 'Go save a million people' – fine, but why would I care about that? But if you have one character who cares about those million people, or a member of that group, and the player can interact with them – if the player cares about them, they will care about the group.
Like Tali from the Mass Effect games? She was the one that taught players all about the Quarians.
Exactly. Having character that represent and embody the larger conflicts at work is what’s going to make the player care – because you don’t care about conflicts, political concepts and nations. You care about people.
What advice do you have for developers wanting to create a new fictional universe?
A lot of times writers are tempted to build worlds that sound great on paper and would be great to read about, but not necessarily great to play in. They’ll have this really complicated backstory, but once gamers step into that world what do they actually do? What do they care about? If it’s all just theoretical concepts and history, that’s not going to be very helpful.
Another thing is, I’ve played a few new IP recently where I wish they wouldn’t take their world so seriously. When you’re being introduced to a new world, you’re having all these new names and concepts thrown at you, but if it takes itself too seriously, there’s no charm. All these efforts are made to make the player learn about it, but not to make them like it.
A lot of times writers are tempted to build worlds that sound great on paper and would be great to read about, but not necessarily great to play in.
If you’re too focused on the lore and the information flow, you might forget that the player needs to enjoy this as well. Enjoying the world is different to appreciating it; you can have a very complex set-up that makes the player go ‘hmm, that’s an interesting concept’, but not necessarily have fun with it.
Try not to make everything different, because then you end up with something that’s completely alien. There is a temptation to go in and completely revolutionise everything, make your fantasy – for example – different to everything that’s come before, but that is both difficult for your team and off-putting for your audience. Start with the familiar and move on from there.
You can read our How To Build A Universe feature here. More interviews will be uploaded throughout the week