Am I upset that Source Code only made 15 million dollars on opening weekend and finished behind an irritating poorly animated bunny? Of course! I wanted the movie to make a bajillion dollars and prove to Hollywood that spec scripts can make good movies too, especially spec scripts as good as this one. It didn’t happen but I can still take comfort in the fact that critics enjoyed it, which is by no means a guarantee with sci-fi.
But that’s not why I’m writing today’s article. I’m writing today’s article because you guys have said you want more script-to-screen comparisons, an examination of the changes made from the original spec to the final shooting script, so we can try and discern what happens during the development process and if that process ends up helping the script or hurting it. Since there’s no script I know better than this one, I thought it’d be a great script to start with.
Before I get into the specific changes, however, let me give you my general reaction to the film. I liked it. Quite a bit actually. I thought Jake was solid. He has a strange quality as a leading man in that he definitely FEELS like a movie star but is missing that – I don’t know what it is – but I guess “swagger” describes it – the thing that separates guys like DiCaprio from the rest of the pack. You want to hang out with Jake. I’m not sure you want him to save the world for you. Still, I thought he worked. As for Jeffrey Wright – look - I know he totally overplayed the part. But I loved it. I love him as an actor and think he has such an interesting delivery that even when he’s hamming it up, I still believe it.
As for director Duncan Jones, I thought he made some good choices and some not so good choices. Someone else pointed out that the movie starts with sweeping shots of Chicago and the train Colter wakes up in. These shots are beautiful to look at but they were totally wrong for the opening, which is supposed to be our character waking up in a strange place with no idea how he got there. If we’re to feel the same way, shouldn’t that be our first shot? Him waking up? Not a $50,000 helicopter shot? It would be like in Buried if we showed sweeping shots of Iraq for three minutes before cutting to Ryan Reynolds in a coffin. Also, I’m not sure I liked the bright bubbly feel of the train sequences and was hoping for more of a cold steely tint, like that of Inception. I understand Jones wanted to clash the train scenes with the dark dreary feel of the source code chamber but that vibrant look just didn’t mesh with the tone of the story.
Having said that, Jones really kept the story moving and, even with the script changes, was dedicated to the spirit of the script. This was close to what I read on the page. And that doesn’t happen all the time. Anyway, here are five key script changes made from the spec script to the shooting draft and how I feel they affected the final product.
CHANGE #1 – 17 MINUTES TO 8 MINUTES
In the script Colter has 17 minutes on the train. In the movie he has 8. At first I didn’t like this change. It felt like too tight of a time-crunch. But the more I thought about it, the more it made sense. On the page, 17 minutes can be fudged to seem like it’s going a lot faster. You just cut out the boring parts. Onscreen, in this particular situation where we’re staying with Jake the whole time, that’s harder to do. Therefore, 17 minutes would’ve felt like a lifetime. 8 minutes is still a little too short for me. 11 minutes would have been perfect. But overall, I think the change worked.
CHANGE # 2 – THE GIRL
By far, the biggest change in the script was making Christina someone Colter knew of ahead of time, as opposed to someone he’d never talked to before. When I first read this change in Billy Ray’s rewrite, I was so pissed off I actually threw the script down and stopped reading. I feared that they’d completely ruined the story with this needless change. Here’s my problem with it. I’m of the belief that you want to make things as difficult as possible for your main character. If things are easy, you don’t have a movie. So for her to just be in his back pocket from the second he wakes up….it’s too easy. I got the feeling that if he’d asked her to marry him right then and there, she would’ve said yes. And how interesting of a relationship is that? That said, I think I know why they did it. And it goes back to the aforementioned time change. In the script, when he had 17 minutes, winning over a girl who doesn’t know you is somewhat conceivable. In the film, when he had 8 minutes, that conceit becomes infinitely harder to believe. He would’ve had to use every single second of those eight minutes before she trusted him enough to do the things he needed her to do. Therefore their only choice was to make it so that she already knew him, so he doesn’t have to use that time to convince her who he is every time he goes back into the train. He can just jump straight into the action. I don’t like it as much as the way the script handled it, but I understand the change.
CHANGE #3 – 120 pages to 90 pages
(note: I don’t know the actual page length of the shooting draft. I’m just going by the 1 page = 1 minute of screen time rule). Now you know me. I’m Mister “Keep your script to 110 pages MAX” Guy. The reason the 120 pages in Source Code never bothered me though was because each section of Source Code is its own ticking time bomb. It’s always a race. So you never feel time dragging. That said, there are a lot of advantages to cutting the script down to 90 pages. The most obvious is that less pages equals a lower budget. Each day you shoot is tens, maybe hundreds of thousands of dollars. So cutting even a single page saves a lot of money. Yet another reason to keep your screenplay lean. Also, thrillers just play better at shorter lengths, especially contained thrillers like this one. Finally, this particular format of movie, where you’re repeating actions, can grate on the viewer if not done right. Go watch the movie “Vantage Point
” to see what I mean. People don’t like to go backwards in a story so it’s best to err on the side of caution and keep things as lean as possible. This movie is supposed to move so they made sure it moved. So overall, I think it was a good change. CHANGE #4 – MORE UPBEAT
I heard Duncan Jones talk about this in an interview – how he wanted to add more humor and a generally upbeat tone to Source Code. I’m usually okay with this. Too many writers drown their scripts in hopelessness and despair, squeezing the life out of their story page by endlessly depressing page. But Source Code’s thriller elements were all perfectly plotted, keeping our focus away from the fact that this was, indeed, a dark story, and instead on the main character’s tasks, which were to find the bomber and find out who Beleagured Castle was. Also, the relationship between him and the girl provided that necessary spark to offset the darkness. That was our “upbeat” storyline. Where I really had an issue with this “happier” approach was the “Here’s ten bucks now go do your comedy routine” scene at the end. You can always tell when an idea hasn’t been fully integrated into a script. The writer tries to squeeze a thin setup for it in early (“You know that guy. He’s a comedian!”) then a full 90 minutes later with no other insight into that character or that situation, we get his Last Comic Standing climax. It’s a total cheat and it felt forced as hell. I’m not saying it couldn’t have worked. It just needed a lot more setup. Overall, I would’ve preferred a darker feel on the train, like the script, so I wasn’t thrilled by this choice.
CHANGE #5 – THE ENDING
Personally, I thought the ending in the original draft was perfect. There’s a couple of reasons for that. First, the images that Colter keeps seeing between the train and the chamber are discussed in the dialogue so as to cue the reader in that they’re an important plot point. Colter asks Goodwin “What am I seeing? I’m seeing something after the train blows up.” This “middle time” is what helps us buy into the ending where they walk into the real world. But in the movie, we only SEE these images. We’re never informed that they’re supposed to be important. As a result, they just come off as a cool visual thing. Second, the ending in the spec draft was simple. They walk off the train, he checks his watch, it’s past 8 minutes, and he’s still alive. In this new ending, there was too much going on, four endings to be precise. We have the comedy freeze, which I admit would’ve been a dark cool way to end the film. Then we have the post freeze, where they realize they’re still alive. Then we have the real world text, where Colter informs Goodwin that he’s still alive in the Source Code. Then we have the “fate” finale, where Colter and Christina go to Grant Park and look at the mirror bubble exhibit. The thing is, I thought each of these endings worked in their own way. I enjoyed all of them. I just didn’t think they worked together. It had a bit of a Steven Spielberg “too many endings” effect that gave you that uncomfortable, “Shouldn’t this be over already?” feeling. So for that reason, I don’t think the ending in the movie worked as well as the ending in the spec script.
So what can we take from all this? Well, changes are going to happen in any script. But you know, they really didn’t change that much in Source Code. And that’s a testament to how well the script was written. This was more “rearranging the deck chairs” than “building a new ship.” And I think that’s why it ultimately worked - because the core elements of the script were always in place. Guy must keep going back in time to find a terrorist bomber before he sets off a much bigger attack. Complications ensue when he starts falling for the girl who helps him. The pace moved. The acting was good. Yeah I liked Goth Christina (from the spec) better than Peppy Christina but Michelle Monaghan is so beautiful that I quickly forgot about her. I really liked the added scene where they go out to the van and (spoiler) Derek kills them. Made him so much more evil and his downfall so much more satisfying. Is it the movie that the original script promised? No. But it was close enough.
What I learned: Whenever you rewrite a script, you’re adding new elements to each draft. Remember though, that while you may be on the 7th or 8th draft of your script, that new element you just added? It’s only on its 1st draft. If you don’t rewrite the script a few more times to get that element into its 4th or 5th draft, it’s going to stick out like a sore thumb. And that’s what happened here with the “Last Comic Standing” ending. It needed a few more beats during the story to really sell it. But they squeezed the element in at the last second and were unable to put it through any drafts. Hence the forced feeling of that ending.