Special Topics In Gameology is an in-depth look at a specific corner of the gaming world, in miniseries form. For this edition of the feature—Funny People—Anthony John Agnello interviews a few icons of video game comedy. The series debuted with Leisure Suit Larry creator Al Lowe, followed by You Don’t Know Jack head writer Steve Heinrich. This week: Portal 2 writer Erik Wolpaw.

Video games are often iterative beasts, evolving through multiple sequels. A game comes out, the players react, and developers go back to make things better with Game No. 2 (and sometimes No. 3, and so on). That process isn’t an ideal environment for comedy, though. A joke doesn’t work as well the second time around, even if it’s been polished since the first telling. It needs to take its audience unawares to work, and it needs precise timing. It takes craftsmanship to make a game that’s funny, but a miracle to make a sequel that’s just as good.

Portal 2 achieved that miracle. Last year, Valve’s sequel to the hugely popular Portal returned players to the subterranean depths of Aperture Science, where a series of “tests” present increasingly complex spatial puzzles. Amid these sci-fi trappings, there’s a core of humane comedy, embodied in the first game by the sociopathic supercomputer GLaDOS (voiced by Ellen McLain). Portal 2 added the idiotic bot Wheatley (Stephen Merchant) and visionary Aperture founder Cave Johnson (J.K. Simmons) to the cast. The Gameological Society talked with Portal 2 head writer Erik Wolpaw about starting with a clean slate for the sequel, steering clear of memes, and his work in the early 2000s at the satirical game-criticism site, Old Man Murray. (Note: This interview discusses the ending of Portal 2.)

The Gameological Society: Walk me through the process of writing Portal 2. What was the very first step in writing the game?

Erik Wolpaw: First step was deciding what characters were going to even be in it. We thought originally: If Portal is a franchise, what is it? Can you do portals forever? We had this other puzzle mechanic called “F-Stop.” We thought that maybe the thing that ties Portal together as a series is that they would all be puzzle games, somehow involved with Aperture Science. So what’s the story there? Let’s not bring back GLaDOS, let’s not bring back Chell! It was going to be set in the ’50s, and you were going to be using this F-Stop mechanic in Aperture before GLaDOS had taken over, and Cave Johnson would be there. It seems obvious in retrospect that making a game called Portal without any portals is insane. And it quickly became apparent that people wanted GLaDOS back.

Gameological: Would Portal be Portal if it weren’t funny? How does humor define Portal?

Wolpaw: That was a given straight from the beginning. Portal 2 would be comedic. That was an axiomatic concept when we started designing. I never want to say “never,” but I can’t imagine sitting down to write a Portal game thinking it’s going to be re-imagined as a serious military shooter or something.

Gameological: What’s the first challenge in writing jokes that require your audience to participate rather than just react?

Wolpaw: You have an idea, but then you have to make sure that you sit with the animators and the level designers and work it out so that the idea works environmentally. We tend to over-record and over-write just in case something just isn’t going to work out once we actually try to wire it up. We’ll have five ideas, and two or three of them end up actually working in practice. There’s stuff that we record, that we think is gold, and it turns out to actually be pretty funny, but then the level gets cut, or it runs too long. Then there’s stuff that plain doesn’t seem funny to us after we record it. Have you played through the game?

Gameological: Yes.

Wolpaw: GLaDOS’ final speech where she deletes Carolyn used to be much longer. Three times longer. We thought it was interesting and funny when we wrote it. It was our last session with actress Ellen Mclain, and when the energy is that low, you know it’s because the writing isn’t good. We started rewriting the speech on the fly. Jay Pinkerton, our co-writer, was directing some of the other lines and I was madly trying to do rewrites. Jay and I were driving home later saying, “This is the final speech of the game and we still don’t like it!” Finally it occurred to us to just use the speech that’s in the final version. Whether or not you like it, believe me, it’s way better than what we originally had. We had to quickly get on the phone and book a session the next day with Ellen just to get it in the game.

Gameological: It worked out for the best. I can’t imagine GLaDOS’ little soliloquy about killing off this human part of her personality being any longer than it actually is.

Wolpaw: Yeah. We didn’t want to trap you in that elevator staring at her for too long, especially because it’s the end of the game and you’re like, “Let’s go, let’s go! We need to get through this!” Also, it was just overwritten. We cut it down to the core. There were a couple of nice beats in it but they just didn’t work. Sometimes you just get that sinking feeling.

Gameological: How do you set up physical gags without making the player feel like they’re just jumping through hoops? There’s that great little joke at the beginning of Portal 2, when Wheatley asks you to speak and the game’s just telling you what the jump button is. How do you keep things like that from weighing down the player?

At no point were we ever thinking, “Aw, we need to make Chell talk!”

Wolpaw: If you’re somebody who plays games a lot, that’s a joke that works good in the fiction but it also works as a meta-joke. Those ones are pretty successful, I think. Then there are physical comedy bits in the game that you’ve seen in television and movies a bunch of times, but that you haven’t seen in games very often. A good one: When Wheatley is saying, “Catch me, catch me, catch me!” when he’s jumping off the rail. That wouldn’t work in a movie now, but being a part of that sort of classic comedy scene, being an active participant in it, rehabilitates it a little bit.

Gameological: What does having a silent protagonist offer you in the writing process?

Wolpaw: This is reductive, but there are two types of comedy. One kind is where there’s a straight man in a world gone mad and the other kind is where there’s a crazy person stuck in a normal world. We’re definitely in the “straight man in the world gone mad” category. You’re the straight man. You-slash-Chell are the straight man. Straight man in a movie tends to get the straight-man lines. It would be strange to have a silent character on film. For better or worse, though, games have a history of having a silent protagonist. We get to make you the straight man, but we also get to cut out all those straight lines.

Having people stand around while you talk is tough in games, so if we can actually remove one whole character’s dialogue, that saves us a bunch of time for all the other characters. At no point were we ever thinking that, “Aw, we need to make Chell talk! This game is lacking in personality!” She probably can talk, though. In our minds, she can talk. She’s just pissed off the whole time and is refusing to dignify any of the things going by speaking about them.

Gameological: The classic Valve stoicism.

Wolpaw: Classic single-player characters, at least. We’re not against letting them speak. The Left 4 Dead characters are all little chatterboxes. After setting the precedent for the Portal series, we felt that it would have been odd to have Chell talk. Once she’s talking, you feel the need to have her comment on everything that’s going on.

Gameological: Why do you meet a friendly face at the beginning of Portal 2?

Wolpaw: We wanted it to feel like a whole new experience, and we also wanted people to be able to play 2 without having played 1, so Wheatley acts as a guide for newcomers while he also creates this abbreviated training arc enjoyable for returning players. With Portal 1, we had this trick that we were able to pull, a trick you get to pull once and never again: We didn’t advertise that it had a story. No one knew what was going on. We were able to start with this very spartan environment with these weird announcements that were kind of funny. That mystery is eventually broken about two-thirds of the way through when you realize that there is a story going on, but that cat’s out of the bag.

Gameological: Portal 2 is a lot more sentimental than its predecessor. There was certainly sentiment in the original, but it was cut with a bitter, sarcastic edge. This time out, there’s a lot of depth to the characters you meet. You feel for them, you even empathize with GLaDOS.

Wolpaw: If we could have just done the exact same thing, believe me, we would have done it. It would have been a lot less work. GLaDOS was basically two notes in Portal 1. We couldn’t hit those same notes, but we also had this longer puzzle experience, so since you’re going to be spending so much more time with these characters, we had the time to make them a little bit more interesting.

Gameological: Way back in your Old Man Murray days, you did an interview in 2001 and you said: “Good game plus crappy writing is probably still going to equal good game.” Is that still true?

Comedy gives you something little to digest in chunks. You get these micro-payoffs.

Wolpaw: Yes. I’m not saying good game plus good writing doesn’t equal better game, but if I’m—I don’t want to give any examples now that I’m in the industry. That would be bad form, even though I’m saying it’s a good game. There are good games out there that, when I play them, I’m skipping the cutscenes, and I could care less what’s going on, but I’m having a blast. There just isn’t an analogue in movies for me. Maybe when I was 15. If there was enough stuff on screen blowing up, then that was enough for me to get through, but now I can’t watch a movie if the acting, story, and writing are bad. If it’s a fun game, though, I can look past anything.

Gameological: As the writer, you want people to care about the story, but there are a lot of people who, like you were just saying, sit down to play, and they can’t get past those story sequences fast enough. They just want to play. What’s the secret to getting them invested in your story?

Wolpaw: Giving them something moment to moment. Comedy helps. Comedy gives you something little to digest in chunks, and if you’re doing it right, you get these micro-payoffs. There’s something funny happening. Then on the macro level, those little jokes are building towards some bigger joke or story point. You bring them along by making every piece of dialogue a reward in and of itself.

Gameological: Incidental dialogue in games, the sort of circumstantial lines busted out intermittently by characters like Portal’s turrets, can get annoying fast. How do you go about writing lines that players are likely going to have to hear again and again and again?

Wolpaw: You try to keep them short and you try to keep them—“not clever” is a good phrase. If there’s too much cleverness, if the line is too complicated a thought, you don’t want to hear it more than once. The defective turrets in Portal 2 only have a little repeated dialogue, and you have to wait a while to hear any of it. The regular turrets, though, only need to say one or two things. Short and not super-duper clever is our recipe for anything you have to hear a lot.

Gameological: When you were writing about games rather than writing games 10 years back, there was still a great deal of space between creators and their audience. That vacuum has pretty much disappeared. Now there’s no way to tune out the audience. How does audience expectation affect your writing process?

Wolpaw: Going into Portal 2, you know the audience is expecting something surprising. You know going in that you’ve lost something. You can’t recreate the surprise of the original. You know that people are burnt out on the memes—the cake. You decide that you’re not going to pander to that which would just be sad and make people angry. I don’t know if we would have talked about cake anyway, but seeing that people, for better or worse, made that their own we figured we wouldn’t go back to that. They own that now.

Gameological: Comedy sequels especially are difficult. Jokes aren’t funny the second time around. How did you avoid Wayne’s World 2 syndrome?

Wolpaw: By having everyone be passionate about the game. Comedy sequels tend to be cash-ins. They aren’t passion projects like the first ones are. Portal 2 was. You had nine people working on 1, then it comes out, and it’s this huge thing. You have this huge talent pool in the rest of Valve, and suddenly everyone’s saying, wow, I really want to work on the next Portal. That brought a ton of energy to it.

Gameological: Why aren’t most video games funny?

Wolpaw: Comedy writing is hard. Especially if you’re just a single person. Luckily, we had three people on Portal 2. If you look at any kind of comedy show like The Daily Show or Saturday Night Live, they don’t have one writer, they have 30 writers. Comedy from moment to moment is about surprise. If games have any writer, they tend to just have one.

When comedy fails, it fails hard. Bad comedy becomes nothing.

Comedy is also tough from a timing standpoint. A lot of games are made with the writer writing the script then handing it off to somebody who hooks it into the game without the writer being involved. For comedy, that’s death. We do the writing, recording, and hooking up ourselves. We’re all in there huddled around a screen, hooking everything up throughout the process, shaving off a quarter of a second here, half a second there, trying to make sure everything is timed exactly right.

The last thing is that comedy is a bigger gamble. Drama either fails at a base level, passing you by without you noticing, or it fails to the point where it becomes comedy. When comedy fails, it fails hard. It’s unwatchable. It’s unlistenable. Bad comedy becomes nothing.

Gameological: What’s the dream project? If time and money and technology were no object, what would you make?

Wolpaw: Whatever it is, it would make someone bankrupt. It would make no money. It would appeal to me and maybe 100 people.

Illustration by Richard Hofmeier.