Before we begin, why don't you tell our readers a bit about yourself? How did you get hired at Obsidian? What games have you worked on? What are your favorite RPGs, modern and classic?

I was hired at Obsidian out of college, though not the first time I applied. My major was computer science, and that was also my best angle for getting hired. They ultimately hired me as a scripter, which is kind of a hybrid design/programming position.

Obsidian was always my first choice, and that was because I'd played Black Isle games growing up and they stood out to me as presenting a new and exciting kind of experience for players. Finishing Torment, in particular, always stayed with me as an experience where everything had come together and paid off beautifully in the end, and I wanted to work with people who made games like that. So rather than apply to a variety of places, which probably would've been the smarter approach, I elected instead to keep pestering the same company over and over until they hired me.

This was at the time they were starting development on Neverwinter 2, so I spent about a month building a mod in the NWN1 engine that was basically Madden but with elves and monsters. That went over well, and it got me hired.

Josh Sawyer has been the public face of Pillars of Eternity, on Internet forums, social media and in interviews. As such, in-depth public discussion has tended to focus on the game's system design and its setting - the things Josh was responsible for. We actually know remarkably little about the hows and whys of Pillars of Eternity's story, characters and themes. Which is of course the reason we decided to do this interview.

So, before we go on to more specific questions, I'd like to ask you to, basically, show your work. What themes and concepts did you intend to explore in Pillars of Eternity? Which quests and which characters are you particularly proud of and how did they reinforce those themes? In short, what's smart about the story of Pillars of Eternity?

I hesitate to talk a whole lot in very specific terms about themes and deeper meaning because people will start to use it as definitive evidence of this or that, and it deadens discussion. I'll say a bit, but you'll have to excuse my being cagey here. Also in order to answer this, fair warning, there will be major spoilers.

We had two major themes we wanted to work with, both of which seemed natural and important to discuss for this particular story and setting. I'll suggest that people who think one of them is about faith might want to broaden their perspective a bit. Both themes are present in the player's story and at least one is present in each companion's story, though which theme it is varies. White March (taken as a whole) puts a spin on both, but tends to focus on one in particular.

Part of the genesis of the Pillars story in particular was the observation that in most fantasy settings, the gods are taken for granted. You know they're up there on Olympus or in the heavens or wherever, and you have some idea of how your afterlife is going to look, and what steps you have to take to improve your standing in that regard. Characters in these worlds, on some level, aren't quite human if they don't have to wonder about these things. It's a romantic and appealing fantasy to have all of that figured out and to only need to worry about killing your enemies and pleasing your gods and boning other similarly carefree and attractive violet-eyed adventurers, and that's resulted in the prevalence of that kind of setting within the genre. But if you go that route you miss out on one of the best ways to test your characters and see what they are made out of, and you also miss out on a powerful source of relatability that just about every other genre has access to (and futuristic sci-fi often thrives on).

This wasn't an idea that came about immediately, even when writing what would become he final treatment, but when it did, it led to the game story as you see it now.

In early development, a big concern I had was that the player's story gets into a lot of talking about big ideas that are often not grounded in specific, relatable realities. That is, as a player, you don't have the same emotional hook that, for example, The Witcher 3 has with the main protagonist searching for essentially his adoptive daughter. It's a big challenge to solve with a silent, player-defined protagonist. BG II was able to draw on the preceding game and use Imoen (although, oh man did I want to let her rot in prison), which was another tool we did not have the luxury of. I made some attempts at solving the issue in reworking pieces of the player's backstory, but that only got me so far. The bigger thing we did to help develop an emotional core to the story, which I felt was more successful, was in working the themes into the designs of the companion arcs and quests. The degree of success varied from character to character, but when I did a full play-through of the game late in development, I found myself enjoying the game's story most when I was seeing the deeper layers of these characters exposed, and their worldviews challenged. Sagani's finale might be my favorite - I found that scene to be very moving.

The other thing worth mentioning with theme is that I think it's often best, in an RPG in particular, to look at themes as questions rather than as moral suggestions. In a novel, you might have a theme about injustice, for example, and the author's ultimate incarnation of the theme might boil down to "Everyone has a moral obligation to fight injustice, whether they are victimized by it or not." But in an interactive, branching medium, it's better to ask, "Are we obligated to fight injustice even when we are not personally involved?" And then you give the player the tools to make his or her own decision. You show them a variety of perspectives. Then you give them an opportunity to act on their own understanding of the matter, as influenced by everything they've seen so far in the game. I can't help but be amused by people who've expressed concerns that Pillars' story is nihilistic (and that's been a number of people), because that's much more a projection of how they've synthesized what they've seen than it is a reflection of some authorial message.


We know who wrote each of the companions in Pillars of Eternity. Can you give us a run-down of who designed what area and what quest?

That starts to get too complicated to pin down, and honestly I couldn't tell you in a lot of cases. There are a ton of quests, and they often change hands. "Everybody worked on everything" wouldn’t be entirely correct, but that's the gist of it. Everybody worked on Dyrford, then we all worked on Defiance Bay, then we all worked on Gilded Vale, then we all worked on Twin Elms. Sorry, I know that's a shitty answer, but doing it right would mean passing out surveys.

The gameplay of Pillars of Eternity was obviously inspired by the Infinity Engine RPGs. Less obvious perhaps is the degree to which its narrative was inspired by them. How did the story, the places and the characters of the Baldur's Gate series, the Icewind Dale series and Planescape: Torment influence Pillars of Eternity? Are there any direct analogues between the games?

Conversely, are there any areas in which you deliberately did things differently from the Infinity Engine games?

Structurally I think we'd be closest to BGII, in the sense that we have a protagonist with a past that unfolds in quasi-dream sequences over the course of the game while the player pursues an antagonist who holds the solution to his problem. That comes from looking at it in retrospect, though. At the time of writing, those elements came about to fulfill various needs of the form and of plot points we wanted to hit. It just happens we ended up approaching the same problems in similar ways.

We also wanted to open up the game as soon as possible. As soon as you were out of the intro area, we wanted the player to be able to explore and go to places way off the beaten path that might be designed for much higher level characters. That's something you'd find both in the Baldur's Gate series and the Fallout series, not so much in Torment, where you're progressing through the areas more or less linearly.

From Torment and BG II both came a sense that the degree to which a story is personal is a strong influence on the impact of the narrative. I don't think our story ended up being as strong in that regard, but it was a guiding principle, and it's the reason we elected to develop a past life backstory rather than having the player be a complete blank slate. This was another case where the parameters of the setting led us down a similar path to an IE predecessor - obviously Torment broke RPG ground in its exploration of past lives. In our case, past lives were the best tool the setting offered that allowed us to give the player character both a background and a blank slate without having to resort to amnesia, which has become a big cliché since the days of Torment.

The actual specifics of the storyline I felt pretty strongly should not emulate the IE games, best we could help it. We wanted the overall Pillars experience to be both familiar and fresh, with story being squarely in the latter category. We'd never stand a chance of evoking the same feeling of playing an IE game for the first time if it felt to players like it wasn't the first time they'd seen the story. So, for example, if we were going to use past lives, there had to be a fresh spin on it, and that was how the idea developed of the player defining their own backstory as it unfolded. That had advantages and drawbacks, but taking the risk there helped make the story experience unique.

Are there any other games that influenced Pillars of Eternity, like maybe Neverwinter Nights 2 or some other previous Obsidian RPG?

All of Obsidian's games influence us, insofar as the experience making those games has resulted in a lot of lessons learned. For example, a good takeaway from NWN2 was, don't make 70 cinematic dialogue cutscenes in an engine isn't built to make them with stability, ease, and attractiveness. We then applied that lesson to Mask of the Betrayer, but it also influenced Pillars. We didn't have time or resources to build cinematics tools, nor hi-res models that would look good super close, so we minimized the number of cutscenes the storyline would require, and tried to pick ones that would look all right from a distant camera shot - often things with big moving parts or visual effects. It also influenced the decision to include prose rather than to try and convey much with specific animations. You have to plan your project to make the best of the resources you have. That sounds obvious, but some people really struggle with that when they get excited about a particular idea or feature.

In terms of content, I don't think there's much we cribbed from our previous titles or from other games. Often you'll look to recent similar games to see how they solved your problems, but in our case, few of those games existed. Wasteland 2 and Divinity: OS came out during development, but they were different enough that I don't think they altered much if anything of what we did.

Certainly the way we treat companions has been an evolution internally. We explored branching character arcs quite a bit in New Vegas, and expanded on that in Pillars. In New Vegas, I think a good number of the companions could have maybe 2-3 endings depending on what conversations you'd had with them. In Pillars, Kana Rua might be our most extreme case - he had, I think, 6-8 different endings. And Zahua is up to something like that as well now, after White March Part II.


How about non-videogame influences - books, movies, TV shows, comics, etc?

Books if anything, for the most part. The themes we explored, or versions of them, come up frequently in modern and postmodern literature, with varying suggestions (or sometimes none at all) for how we're supposed to come to terms with them. There aren't a lot of books I looked at specifically for this, it's more just the general influence of having read them at some point or another. One that comes to mind is The Crying of Lot 49, which has a much different tone, but there's a kind of feeling you get when you read certain parts of it that was probably the same kind of feeling we were shooting for with Pillars. Maybe a little of Night-Sea Journey, too. I'm sure there are others. Those influences were all pretty oblique - there was nothing we explicitly tried to imitate.

The other thing was, we wanted to do a very active, scheming pantheon of gods. There is a series called the Malazan Book of the Fallen that has a similar level of involvement (probably heavier than ours), and someone referred me to it. Might've been Chris. I have only read a small part of the first book, but I think it was good to see a sort of "proof of concept" of that kind of a setting.

You were employed as Writing Lead on South Park: The Stick of Truth during much of Pillars of Eternity's development. While that sort of multitasking worked well for Chris Avellone in the late 90s when he did Fallout 2 and Planescape: Torment simultaneously, one can't help but feel that it was a bit frivolous of Obsidian to divide the attention of their inaugural Kickstarter title's lead narrative designer in such a way. Why did things turn out that way? Do you think it harmed the game?

Being split between two projects is always a challenge, never a boon. It's also a common reality of development. Staff gets juggled according to the most pressing needs at the company. When we figured out staffing on Pillars,Stick of Truth's release date seemed pretty set. I was working on early drafts of the Pillars story when the publisher change happened with Stick, and that pushed everything on Stick back. There was no practical way to sub in somebody for me on Stick - I was the only in-house writer on that project, and it was a very specialized kind of job - nor was there somebody in a good position to take the narrative lead on Pillars. We just had to ride it out. Nothing frivolous about the decision - it was just practicality. The least of several evils.

As to what impact it had on the game, it's hard to say. I always want more time. As a writer, the more time you have, the more you can rewrite, the better the final product. Not always, but in general. But that kind of second guessing doesn't really get you anywhere, because game development is never ideal. So much of it is learning to make the best of the situation you're in. There's maybe a handful of people in the whole industry who are allowed to hold up their entire project until they're happy with their own writing. For the rest of us, there are budgets, and there are full-time employees spinning their wheels waiting for you to give them a story that's final enough that they can start working on it. You're always behind, hoping what you're cranking out isn't terrible, because you probably won't get a chance to revise it.

In my ideal world, you wouldn't start production on a game until you had a finished critical path script. That's how they do it in the movies (or did...), and it's a good way of knowing in advance if your story is going to be crap or not so you can only produce your best stuff. It's also impractical because it takes so long to write a branching RPG that you can write full-time for the entire length of development and still not be done in time. And the rest of the team needs something to do. It works for movies because they staff up and lay off on a per-project basis, so they're always lean and mean for preproduction. Someday I hope somebody figures it out, though. I think a lot of games suffer because the story was an obvious afterthought.

Josh Sawyer and Adam Brennecke have tended to downplay the significance of any content that may have been cut from the final version of Pillars of Eternity. Do you feel the same way? Is there any cut content you'd like to tell us about?

I'm not sure what the question is referring to regarding Adam and Josh. We made cuts. Some of the cuts made me sad. But they had to be made or the game wouldn't have gotten done.

Two big ones had a substantial impact on the story, although both happened early-ish in production, so the content was never built. One was that we cut the next-to-last level of the game - or rather compressed it down to a single map, which contained little content. (This is Breith Eaman, the prison.) That cut hurt pacing quite a bit. The end came up very abruptly. I'd have loved to spend more time at least doing some more repairs to that part of the story, but that wasn't possible. The time just wasn’t there, and I think I also underestimated the impact. Ultimately, when you are told you have to cut something in the story, you have to be prepared for that and have some answers. In this case, I was able to stitch everything back together so that at least it all made sense, but I'd have liked to have gone back and seen if there was a better way to solve the problem.

The other one was that we wanted to branch the middle of the plot. Some people have expressed frustration at the player's inability to influence the outcome at the ducal palace. Well, originally, we'd wanted the player to be able to do that. But it meant building two versions of the third act, and that's extremely expensive. That cut made me sad, but there was no practical argument to be made for keeping it. It was a clean cut that saved a ton of time and made our schedule semi-workable. Had to be done. Conceivably we might've allowed the player to save the duc without doing a major branching of the story, but even that would've required more time than we had. The game was delayed as it was, so there really wasn't room to add anything. As a developer at the end of a project, I think it's almost inevitable to find yourself thinking "man, the things I might've done with a couple more weeks." You're Liam Neeson at the end of Schindler's List, wishing you could have done more.


Chris Avellone created the initial versions of Durance and the Grieving Mother, but then they were cut and passed on to you and Carrie Patel, under circumstances which remain unclear. Since they're such distinctive characters, the widespread assumption among fans is that most of their writing is still his (you seem to agree), although he's reluctant to take credit for them. Can you tell us your side of this story? What was changed about these two companions and why?

I don't know about "my side of the story." I can tell you the extent of my work on Durance. I wrote (if I'm remembering right) some of his environmental reactivity (like, for example, what does he say when he sees a dragon or the Grieving Mother drowning in a pool of blood), and then all of his banter with other companions, and his interjections into other conversations. I also gave the player an opportunity to call him out on his self-deception and hypocrisy, because it seemed to me that some players would want to, and that they might be more inclined to keep him in their party if they could, despite him being not the nicest guy. I had to make some minor edits to get everything to line up and make sense when his dialogue was pared down for length, but not a whole lot. Chris chose what to cut, and it was fairly clean - there was a layer that could be removed without losing the base of the character. Carrie's work on the Grieving Mother would've been similar, though I'm not sure the specifics.

The cuts came for length. The three limiting factors were time to implement, art resources for the dream sequences, and VO budget. There was a target length we had set upfront for all companions, and we had to stick to it. Otherwise we'd be, for example, voicing maybe one out of every six lines for Durance and the Grieving Mother, and it'd be conspicuously incongruent with the other companions, who had maybe 2/3 of their lines voiced. Unfortunately in this case it meant cutting down characters that had had a lot of research and creative energy invested in them, and there were some good ideas there that it would've been interesting to explore. It was a shitty thing to have to do, but we'd never have been able to implement the original versions in time to ship.

More interesting than cut content perhaps is content that never even made it off the drawing board because you didn't think there was a chance that it wouldn't eventually get cut. Is there anything you wanted to add to Pillars of Eternity at an early stage but backed away from due to it being too ambitious?

I mentioned this at our ComicCon panel, but I was hoping we might do low-INT dialogue, and I had a pet idea that we would also have low-INT journal entries, since our journal is written in first person. That's obviously very expensive, because depending on how far you take the joke, you're potentially doubling up your word count. But I can still dream.

Early on, I wanted to lapse time during the middle of the game. We thought about showing the player the world in multiple ages, which would've been awesome, although probably also disorienting for players who are still trying to get a handle on the setting. I'll get it into a game eventually.

We know that during the early development of Pillars of Eternity, there were multiple competing pitches for its storyline. Obviously, your pitch won. Can you tell us more about those alternative storylines, and why they didn't make it?

In retrospect, I don't think we would take that approach again. It's very difficult to gain momentum because everyone has a dog in the fight, and nobody is quite as excited for your idea as they are for theirs. And actually we didn't go with my pitch, either. Everything was kind of a stalemate at that point - there was no consensus favorite.

The good part was that all the pitches had something interesting about them. So after that I worked with George Ziets on a new draft that built on some of the favorite elements from all of the pitches. That got us closer, and there are a number of key elements there that made it into the final story. Thaos and the Leaden Key existed in a form in that draft as the antagonist, and the baby plague was in there, and some of the key events and locations. But the reception on that draft was still a little tepid. So there was one more iteration after that. That was where the twist about the gods was introduced, among other things. Was the reception any better for that one? Maybe, slightly. But at that point it was time to write the game, and I thought I could run with it, so that's what stuck.

During the Kickstarter, several characters were introduced that didn't make it into the final game - Cadegund, Forton (who eventually returned as Zahua in the expansion), and an unnamed Orlan cipher detective. A more masculine-looking Devil of Caroc was also shown at one point during the game's development. Players have found evidence of these characters in the game's files (see this thread on the Obsidian forums, for instance). What happened to them? What were they originally supposed to be? Is there any chance of more of them coming back the way Forton did?

Cadegund and the orlan detective became Durance and the Grieving Mother, respectively. The former two became redundant, class-wise, after Chris pitched the latter two as the characters he wanted to work on. (The goal was to eventually have a companion for every character class.) To my knowledge, hardly any design work had been done for Cadegund or the orlan cipher at that point, beyond their appearances, so there wasn't much that was cut. Josh may have had some thoughts as to their backgrounds, but I never saw it on paper.

Given that, it was maybe a little surprising that those "cut" characters (Cadegund in particular) seemed to have fans among the public. But they looked like a couple of badasses, so I can see why people might wonder why we'd cut them. We haven't talked about it here, but I wouldn't necessarily rule them out in a future game. They have their fans internally here, too.


If the game's characters changed over time, it stands to reason that other things changed as well. In general, how did the storyline of Pillars of Eternity evolve over the course of the game's development? Were there lots of iterations or did you get things mostly right on your first try?

I think that's mostly covered above. Most of the changes occurred in preproduction. Once we got into production, the story did not change very much - there simply wasn't time to iterate on it. The cut to the next-to-last level was made in production, but before we'd begun working on that region.

The Raedric's Hold questline was one of the most well-regarded ones in the game. A multi-area set-piece, lots of choices, lots of approaches, multiple actors with their own agendas. Most of Pillars of Eternity's quests weren't as elaborate and it really stood out. As the designer of Fallout: New Vegas' remarkable Beyond the Beef quest, you're no stranger to that sort of thing. Can you tell us more about the making of the Raedric quest? Can we expect to see more quests like that in the future?

We all liked that one internally, too, and you'll see a similar paradigm for a couple of our larger quests in White March Part 2.

In my position, when I'm writing, I'm usually focused on the critical path, and I don't often get a whole lot of time to help flesh out side quests, but this was early in production and there was some opportunity to work with the level designers in conceptualizing the storyline for the quest. So I think that one, on the narrative side, began with the image of Raedric on his throne, half-crazed, after slaying his wife and child. We wanted strong ties to the central crisis of the game, to help the player begin to understand the stakes of the plot.

Then the level designers ran with the quest design and added some nice touches to the plot there as they went - Kolsc, the would-be usurper you could prop up or kill off, for instance, and the NPCs inside the keep that give insight as to Raedric's state of mind. Olivia Veras, who does writing as well as level design, did most of the writing for the quest and I thought executed it beautifully. Because it was early, that quest got a lot of scrutiny and a good amount of iteration, and I think it came out so well as a result of strong collaboration among the team.

The thing to know about a quest like this is that it is expensive. Introducing layers of complexity like multiple start and end points inherently takes much longer to implement, and generates far more bugs (geometrically, I'd say) than a simpler, more linear quest. There is a quest in Defiance Bay with a wizard masquerading as a drug dealer, and one in Dyrford about someone hiding from the local crime family, and those quests both had those types of complexities about them, and both required substantial iteration to make serviceable, and had to essentially be scrapped and redone multiple times. Those quests then start to eat into time that could be spent polishing other content, and it's not always clear-cut whether the quest will be worth the trouble. Over the course of production, we've gotten better at making that judgment and identifying red flags, but it's always a little risky. All of that is to say, we love doing that kind of quest - maybe too much - but if we tried to do every quest like that, the game would be a disaster, so we try to choose our battles based on the ones we think will pay off the most.

On New Vegas I worked on Beyond the Beef but I also did one of the vaults - Vault 11. Vault 11 was a relatively linear, exploration-based quest and probably had half a dozen bugs reported total during production. I probably spent triple or quadruple the amount of time I worked on Vault 11 working on Beyond the Beef. So much of it depended on nondeterministic systems (AI, scheduling, pathing...) that the whole thing was a nightmare to debug. And when the game came out, it was interesting to see that Vault 11 got a lot of public praise, but hardly anything was said about Beyond the Beef. I'm glad people like it now, but at the time it did teach me that there are many ways to make a memorable quest, and some are harder than others.

Another well-regarded aspect of Pillars of Eternity is the writing in the Council of Stars questline, when you communicate with the gods in their domains. Really top-notch, imaginative stuff, in my opinion. Can you tell us more about the making of that? What were your inspirations?

Glad you enjoyed it. That was Carrie nailing the writing. Some of it, I think, was that the assignment had few boundaries and gave her the freedom to spread her wings and show off a little bit. You can only write so many down-on-their-luck peasants with missing family members before they start to run together for you. But here, the assignment was, the player is going to pray and have a dreamlike vision in which the gods speak to and attempt to manipulate him. We weren't going to build the vision environments (although we'd have loved to if we could have), so there was absolute freedom to use whatever imagery best set the atmosphere for a given god. It was really the biggest opportunity we had in the game to leverage the prose. And the gods themselves only existed as paragraphs in a lore document, so Carrie was able to give them whatever voice she cared to envision for them.

We structured it so that the first encounter with a god was more mysterious - you don't talk directly to the god and it's more about having the player ask, what is the god trying to get me to do, and is there some hidden agenda here, and what are the moral implications of going along with this particular god or group of gods? Then upon returning to the god, the god would speak directly to the player and explain how the quest illustrated their values, and the player would have to answer for how he chose to complete the quest.

The one outside inspiration for this was that we sat down together and went through the Myrkul dialogue from Mask of the Betrayer, which George Ziets wrote and is to this day one of my favorite dialogues in an RPG. We looked at it specifically because I felt it was a good example of the sense of grandeur one should feel in speaking face-to-face with a god. And I felt Carrie succeeded in bringing that scale to the gods of Eora.


Pillars of Eternity's endgame revelation isn't much of a spoiler at this point.

Okay I'm going to stop you there. It's a HUGE spoiler! It's not a spoiler if you have finished the game, or are RogueZombie and feed off of spoilers like some kind of spoiler vampire. For everybody else, there are fucking spoilers below. But you probably should know better than to be reading this right now anyway so I don't feel too bad for you.

Some players think that you revealed the true nature of Eora's gods at too early a stage - that a discovery of such metaphysical significance would have been more appropriate in a sequel, as the finale to a grand arc. What's your response to that? How do you up the ante in future games after revealing what appears to be the setting's greatest secret?

We could've done a trilogy of games where we built up the existence and importance of the gods and then revealed - probably at the 2/3 point, Empire-Strikes-Back-style, that they're not really gods. That's a perfectly good way to do it. In that scenario, the Leaden Key probably remains your antagonist through at least the first two games.

We chose to do our story differently, and that was with purpose. Letting the cat out of the bag in the first game allows the story to evolve more quickly - the revelation can set new things into motion and raise the stakes for future titles. There can be additional interesting turns, should the sequel gods smile upon us.

Also, this is not like D&D (or any modern blockbuster franchise), where there is a whole universe and you are limited in how far you can fuck shit up because other people have to be able to use the setting for their bullshit stories about C3PO's red arm or whatever. Granted, Feargus probably won't let me destroy the universe, but I am fortunate in that I'm not as obligated to hold back as I would be with a licensed title. I'm allowed to tell a story in which the revelation that the gods aren't real is just the start of something bigger.

If the criticism is that we didn't do a good enough job making the stakes clear by ensuring the player had strong familiarity with the world and its gods before he's in there talking to them and learning their secrets, I think that's valid. In other words, it's not a big deal that there are no gods because you as the player barely knew them anyway. Ideally you want to develop that in your side quests, and we did that to a degree, but I don't think we covered enough of it, or made a big enough deal about religion and the pantheon in the first two acts. Completely new fantasy settings bring with them a host of writing challenges, none of which I understood when I was working on the story at the beginning of the project. Doing it again, I think I'd want to spend more time mapping out where and how the player should become familiar with key aspects of the setting, and make sure the pacing on that was good and not front-loaded, so we're ramping up at a more natural pace. It's a tall order, though, any way you slice it.

There have been a few recurring criticisms of Pillars of Eternity's narrative which you've probably heard of. From beginning to end, the major ones are:

1) A lack of motivation at the beginning of the game ("Okay, I'm a Watcher, so what?")
2) The stronghold - its lack of content and its disconnection from the narrative
3) A loss of inertia in the game's third act, due to Twin Elms' isolation from the conflicts that drive the main storyline and its lack of a strong central arc

Josh Sawyer has already issued a mea culpa over some of these, but I was wondering if you had anything to add.

I'd have liked to do more with the insanity end of the player's affliction to help push immediate motivation. In an ideal world I'd have wanted to push it almost into Eternal Darkness territory, trying to make it more present and memorable. But that would've required quite more time and a greater resource investment.

It's a surprisingly difficult balancing act to motivate a player along the main quest of your game. One reason for that is that different kinds of players resent that kind of push to varying degrees. Players who self-identify as explorers may hate you for making them feel rushed if you employ a ticking clock device. (That was one major source of criticism on Mask of the Betrayer re: the spirit-eater system.) I think that's why you see a lot of open world games dispense with the ticking clock entirely and just decide it's not a big deal if the player wants to forget about the main plot and goof around. New Vegas took criticism for its early-game motivation, too(do I really care about getting revenge on this guy?), but I think it's easier to forgive in a systemic sandbox engine where the distractions and side paths are the stars of the game. For us, we did have to be careful not to push so hard that we didn't give people an opportunity to explore this new world on their own time, but I think there was more that could've been done in terms of motivation.

My first story pitch actually had a climactic defense of the player's stronghold! But this was (rightly, I think) regarded as too costly and was one of the things that ultimately sunk that pitch. The last time we attempted a stronghold siege was NWN2, and that ended up being quite a lot more expensive to develop than anyone anticipated. It was considered, too, that in BGII, the strongholds didn't have much of a connection to the narrative beyond the quest to acquire them, so I think that was chosen more as the model. Fortunately, the new 3.0 content for the stronghold goes a long way toward making it feel more integrated into the game. I think people will enjoy it.


As for the third act, the truth is, we had fewer resources to invest in it than we did the second act. If I had to guess, I would say it got half the resources that Act II got. I could be off base. And we did have to nix some of the original plans for the plot there. It's one I'd have loved to have spent a lot of time fixing up.

You may have identified the recurring theme of resource scarcity. None of this is to say we couldn't have figured out things better on our own, or didn't make mistakes. I personally made plenty. But in most cases, the biggest thing separating the game from the heights of its potential is a simple lack of time and money. "Writing is rewriting" is a great quote about writing that I read that was written by somebody who never worked on a mid-budget video game. Most of the dialogue in Pillars is first-draft with a cursory editing pass. There was very little time for iterative improvement, especially later in development. You can be an Uncompromising Creative Genius about this predicament and make yourself very miserable and probably never have a game come out because you're just not happy with it yet (and these people are numerous and well documented), or you can learn to accept it, and be grateful that you aren't working on an Uncompromising 5-10 year per game development cycle.

Personally, I would like to see us make shorter games (e.g. 30-40 hours instead of 60-80) where we cut the worst of our content and spend time iterating on the best. But there is pressure from the market itself (or at least perceived pressure) to make longer games so as to justify the game's sticker price with its "value" as measured in dollars spent per hour of gameplay. And I'm not sure if people understand that when you're on a budget, there's a zero-sum tradeoff between gameplay length and gameplay polish. There was some backlash for Stick of Truth, for example, for being "too short" at 12-20 hours. But that was a game where we cut the bad stuff and spent extra time on the good stuff, and I prefer that model. As a gamer, I'm getting old. I'm short on time. I'd rather spend $60 on a 12-hour experience that makes me laugh my ass off than on a 100-hour experience that routinely wastes my time. If any of you are in agreement, be vocal about it, because I think the dollars/hour guys are usually louder. Come to our forums and ask for a shorter, more polished game. If you don't feel that way, shhh, you, shhh.

There are some who claim that Pillars of Eternity's writing is overly verbose - not just compared to modern voice-acted RPGs but even compared to the text-based classics that inspired it. In particular, they claim that its dialogue contains too many "exposition loredumps", an assessment that is particularly surprising considering that Josh Sawyer opined heavily against such loredumps early in the game's development. Do you agree with that assessment? Was there perhaps a deliberate effort to add expository dialogue to Pillars of Eternity, to help introduce people to the new setting? (As opposed to the Infinity Engine games that could rely on a large audience of Forgotten Realms aficionados.)

It's too verbose in many places. The beginning was egregious. I'm to blame for a lot of that. Part of it was that I hadn't written prose in a long time. I found my stride later on. I am very sorry.

A separate, but equally large part of it was the exposition. Nobody likes writing exposition. You feel unclean when you've written it. It's boring and it doesn't advance plot or do anything worthwhile at all. Unfortunately, in this case, there was a lot that had to be conveyed for you to even understand what was going on. You had to know what a bîaŵac was before it struck. You had to know what adra was. You had to know what a Watcher was very shortly after becoming one. You had to know who Glanfathans were and why they would be mad at you for being in their ruins. You had to learn about animancers and the Saint's War and a slew of other things that led to the world being in the state it was in.

Later on in development, we got kind of a hyperlinked tooltip system that explained certain highlighted words when you'd mouse over them. This was used to explain systems primarily, but if I'd have known about the system early on, I think I could've made a lot of the early dialogue cleaner by offloading those explanations into some database the player has to opt into. Wouldn't have solved everything, but wouldn't have hurt.

Let's talk about the backer NPCs and their little stories. Yes, they were clearly marked as optional, but you know, hardcore RPG players always want to go everywhere and read everything. As such, many complained that those NPCs killed the game's pacing when exploring towns. Not to mention the weirdness of having godlikes and other "wish fulfillment"-type characters just standing around all over the place. Surely it would have been preferable to integrate them into the game's actual content, the same way Torment: Tides of Numenera is doing. Why did you end up doing things the way you did?

The biggest regret, I think, is that we didn't do a better job telling the player immediately that they were optional. To my knowledge, we didn't have feedback that certain player types obsessively click on each and every one until after we released. We probably should have anticipated it, but it was just one of a huge number of things we were concerned about at the time.

Integrating them into content would've been extremely challenging given the sheer number of backers who created characters. The backer-designed character thing was also something that was conceived (probably frantically) during the Kickstarter process. I don't know how much thought was given to practical concerns of implementation at the time. (I wasn't on the project yet, so I couldn't say.) Eventually we had the idea to connect the backer NPCs to the Watcher condition and use that to tell their stories, and we had some extra writing resources available at the time as people were transitioning between projects, so that felt like an elegant solution. We didn't realize then that it was integrated so seamlessly that people would mistake it for being plot-relevant.

I haven't seen what Torment: Numenera is doing, but hopefully they've learned from our mistakes, and I'm interested to see what they've come up with.


Some people have spotted a certain similarity between Pillars of Eternity's animancy hearing and Neverwinter Nights 2's trial - the way they both tie off the games' second acts with a singular outcome, no matter what you say in them. Although I don't think it's a perfect analogy, there's a certain kind of RPG player who can't help but resent any sort of railroading plot device. Would you have added more tangible (ie, non-ending slide) consequences to the player's choices in that section if you could have?

Yes! I've covered that above, but yeah, man, we wanted to branch the whole plot right there. We ended up doing as much as we could - the player can change the duc's decision regarding animancy and there is reactivity immediately in the riots and then in the game ending, but the duc is fucked either way and that was too bad. In a hypothetical sequel, I could still see the outcome of the hearing having longer-term implications, so we may not be done with that hearing yet.

Pillars of Eternity is the first RPG of its kind to be developed in over a decade. Historically, Obsidian has strived to be an AAA developer, and the game must have been an anachronism to you guys in many ways. I imagine there were a lot of lessons you had to re-learn - or perhaps, un-learn. Can you share some of them?

Narratively, in a lot of ways Pillars evolved naturally from everything that came before it. The fact that the game was based on an older aesthetic wasn't often constraining to my work, and the only practical adjustment I had to specifically make was that I felt it would be a good idea to use prose, knowing that we wouldn't have cinematics or many animation resources. So I did have to remember how to write prose, and that took a while. (Working on the Pillars short story about Edér more recently ended up being a great exercise for that, and I regret not attempting something like that on my own sooner just to shake the rust off.) But in general, we were able to play with a lot of new ideas - dynamic backgrounds, personal reputations, that sort of thing - because we're building on systems and ideas that we're deeply familiar with.

Looking back at the development of Pillars of Eternity, is there anything you would have done differently? Stuff you would have wanted to flesh out more?

Tons of stuff I'd have liked to flesh or polish more, resources permitting. Other than what's already been discussed, I think the biggest thing on my wishlist would've been an opportunity to do a closed, confidential alpha test of the entire game, and have gotten feedback on the story with a block of time to iterate on it afterward. Not so much to make sweeping changes, but more to find out, where are things unclear, where are we losing people with the main story, that sort of thing. That kind of iteration can make a huge difference in the end.

On the other hand, you look back at the project, and at the decisions that were made, and you realize that they were usually made with good reason. The toughest ones are the cuts, but there wasn't any getting around them. Other times there might be more investment on one type of content at the expense of another, but there was always a good argument to be made for either side. That's not necessarily a common thing you can say about a project in this industry - that there wasn't some major fuck-up that cost a lot of time or made some aspect of the game a disaster - so I would say this one went well overall.

If anything, I think with the benefit of hindsight, we probably could've dialed back our Kickstarter stretch goals. They achieved their purpose at the time, which was to drum up enthusiasm among the backers, but some of the bigger goals probably cost quite a lot more to develop than they brought in - the second city, the megadungeon, and the expansion come to mind. Smaller scope would've afforded us more polish time.

Let's talk about The White March. Some of us are waiting for the second part to release before we play the entire thing. Others have dismissed the expansion entirely as a pointless dungeon crawl in an already combat-heavy game. They would have preferred a Mask of the Betrayer-style "sequel-pansion" that makes a clean break from the base game with an ambitious new story. So my question is, how would you sell The White March from a narrative standpoint? Does it address any of the base game's criticisms?

Well I've taken enough time to respond to the interview that the thing is out now and can speak for itself, to a degree.

For the story, we knew the expansion would be released in two parts, and we didn't like the idea of telling two independent 10-hour stories. We wanted to do something with epic stakes, and you have to build those or they don't feel believable. 10 hours wasn't enough time. So while the expansion is playable from the middle of the game rather than as a standalone, we did hope to release something that felt big and important and worth the player's time. Our challenge was in making a story that felt complete both at its midpoint and at its conclusion, so it could be split without giving players aneurysms. The idea we settled on was to tell a small, focused, dungeon-exploration-driven story that set something much bigger into motion, and that the player would be the catalyst and thus have personal investment in the plot.

The bummer was having to read reviews about Part 1 that lamented the small, impersonal, unimportant-feeling story. If there were some practical way to have told reviewers to not review the game until Part 2, that would've been ideal from my perspective, but that isn't how the business works. Taken as a whole, I think the work makes a lot more sense and feels much better than when evaluated in serial.

Now that you can play both, though, I think people who enjoyed the base game will really like the expansion, and so far that's been the bulk of the feedback from press and the internet masses.


It's becoming increasingly obvious that Pillars of Eternity is going to get a sequel. The most recent press release for The White March Part 2 has all but spelled it out. Now, we don't expect you to confirm that or offer any specifics, but...what sort of things would you be looking forward to doing in a hypothetical Pillars of Eternity 2?

Hypothetically, I have a few things I would want to play with. I don't want to tip my hand, so pardon the vagueness. One would be having fewer, but far deeper and more interconnected companions - interconnected both with respect to one another and with respect to the overall plot. "FEWER?! FUCK THAT," you say. But everything is zero-sum in this business, and every companion we add takes a ton of time to write and implement. So yes, fewer. But better. More memorable. More like a real group of people. Less likely to be collecting dust in your stronghold.

The other big thing is I would want to do a plot that is more dynamic and player-driven. Pillars is no more linear than Baldur's Gate or Planescape: Torment, but the structure was still A to B to C to D to E. Next time, if there is a next time, I would want the player to have some agency in picking where the plot goes next, and I would want the world to respond to it in big, meaningful ways during the game. Big talk, I know. But I do have an idea for this that pushes dynamism and puts my degree to use (for once) that I would be excited to experiment with.

Some people have theorized that if Pillars of Eternity is like Baldur's Gate 1, and The White March is like Icewind Dale, then a Pillars of Eternity 2 would surely be the modern successor to Baldur's Gate 2, with its "everything goes" cosmopolitan city setting, wacky high level shenanigans, a more personal main quest, and tons and tons of content. Will you guys continue to follow in the footsteps of the D&D classics, or do you plan you branch out in your own direction? Also, do you think there's a chance of Obsidian ever making a sequel that's that big? Even BioWare admitted that they went overboard with the amount of content in their postmortem back in the day. But it's a big part of BG2's status as an enduring classic.

The IE comparisons will always be inevitable, I think. We certainly invited them in our initial pitch. For us, there are aspects of those comparisons that are useful, and aspects that aren't. Production-wise, when we scoped White March, considering budget and timeframe, we were able to look at Tales of the Sword Coast and make some initial projections about how many maps we'd need to create to achieve a certain amount of content. That was useful. It'd probably also be useful to look at the jump from BG1 to BG2 and evaluate the success of various improvements they made to their game, and consider if they would be worthwhile additions to a hypothetical sequel.

But I wouldn't expect or want to use BG2 as a template and say, well we have to add x% of gameplay hours and y companions and z new levels and the story and setting have to be such-and-such just because that's what they did. It would probably be a mistake to try and replicate their success as though our games were identical and the world had stood still for the last 16 years and the cost of development hadn't changed.

Scope-wise, as you pointed out, it's probably overboard. Steam achievement statistics suggest that only about 9% of Pillars owners completed the game. In many cases that's probably because we lost the player's interest, but it also suggests that the game was already plenty big for the majority of our player base. Whereas if we were to invest in depth instead, that's something players get to enjoy whether or not they finish. If we were to end up being bigger, great, as long as we prioritized getting better first.

Creatively, I think it's boring to not try and push the envelope in one way or another, and a fool's errand to try and make a game that tries to trade punches with a formative childhood memory. So, personally, I'd like us to go our own way, taking what lessons we can apply from the IE games, but ultimately making decisions that result in the best Pillars sequel we can manage, based on everything we've learned in our combined experience. I don't think you can make a new classic by emulating an old classic. Or at least I don't think it would work in this case.

My final question may be a bit of a doozy. By most accounts, Obsidian Entertainment has never been in a better shape than it is today. You're no longer at death's door like in the bad days of 2012. You've created your own intellectual property. You've managed to take your place as the successor to Black Isle Studios, creating complex, critically acclaimed PC-exclusive RPGs. The future seems bright.

Yet there are those who look at Obsidian today and don't like what they see. To them, Obsidian has become a play-it-safe, uninspired and po-faced developer, terrified of its "Bugsidian" reputation, and no longer capable of producing the imaginative Planescape: Torment-inspired character-driven metaphysical adventures that seemed to be its bread and butter back in the 2000s - games like Knights of the Old Republic II and Mask of the Betrayer. The departure of personnel such as Chris Avellone and George Ziets hasn't done much to dispel those beliefs. What do you say to those people? Do they have anything to look forward to from Obsidian?

There's always this reductive impulse among the public to try and personify companies based on scant information. We like to root for certain developers or demonize certain publishers or attribute entire games to individual personalities because it allows us to have clear-cut opinions and to debate them with others. Those opinions aren't necessarily baseless - they're often based on observed trends (although just as often they're based on rumors and conjecture). But in my experience, things are usually much more complex, because we're talking about a diverse group of people and trying to nail them down to individual traits.

There are people here who like to play it safe and people who like to take risks. There are funny people and humorless people. We've released inspired games and uninspired ones, and I've worked on both. We've lost strong developers and weak ones, and hired the same. The studio's output is a function of an insanely complex dialectic between all these people, nobody has as much influence as you think they do, and many of the people who had the biggest roles in our successes and failures, you've never heard of.

So I would say this. Whatever omens you're trying to read to determine if the company is on a good or bad trajectory, it's not enough information to make a sound judgment. It's a blind guess. Better to wait for the games, and see for yourselves if you like them, and judge every one independently of every other, because they're all different and made by different combinations of people. Otherwise you're just cultivating confirmation bias.

If you're looking for a reason to hope for good stories in Obsidian's games to come, I have this to add. The best weapon we have - and it's by no means a guarantee - is to hire good people. In the past few years, we've set ridiculously high expectations for our narrative design candidates, and the ones we've hired are truly a cut above. So I can say that I personally am very excited to watch these talented writers grow into their roles and make stellar contributions to our next titles, and to hire more people like them, as we find them. I believe there is a lot to look forward to on that front.