Daily Archives: April 26, 2011

Source Code (Canada/France/US 2011)

Jake Gyllenhaal and Michele Monaghan as commuters – are they a couple? She knows him, but he doesn't know her.

It’s turning out to be a good season for intelligent and well-crafted science fiction/speculative fiction. Following Monsters and The Adjustment Bureau, Source Code again offers a cerebral SF movie with a strong romance element – although very different from the two earlier films.

The only surprise for me was that the movie cost so much. Box Office Mojo suggests $32 million – I wonder how much of that went to Jake Gyllenhaal? Or perhaps it was all those aerial shots of Chicago? I wonder how much of it was CGI? The budget is interesting because this is an independent film with no Hollywood involvement. A Brit director (Duncan Jones), an American story (from Ben Ripley) and a largely French-Canadian crew (in Montreal) worked for a couple of independent production companies and the film has generally been released by independents. Business has been pretty good with $63 million worldwide so far and only three major territories (North America, Russia and UK/Ireland) on the release slate. One explanation for the film’s genesis is that Gyllenhaal was attached to the script for some time without a director being signed up. He saw Moon and suggested to the producers that they should hire Jones.

Plot outline (no spoilers)

Jake Gyllenhaal is Colter Stevens, a US helicopter pilot in Afghanistan who wakes up on a Chicago commuter train not knowing what is going on when the woman opposite (Michelle Monaghan) calls him ‘Sean’ and appears to know him well. Eight minutes later there is an enormous explosion. Stevens blacks out and wakes again in some form of capsule. Is he part of an experiment in a simulating experiences or is he dreaming? He talks to a ‘controller’ who explains more about his brief . . . and sends him back to the same train at the same time. This cycle will be repeated several times.


This fits the ‘speculative fiction’/science fiction definition pretty well. The plot depends on the possibilities of developing technologies concerned with direct connection to electrical impulses in the brain. In that sense, the ideas are familiar from decades of SF. Many SF film fans have complained that the film isn’t ‘original’ in terms of its ideas and I’m sure they are right. The ideas could be argued to be ‘Dickian’ in origin – i.e. similar to those of Philip K. Dick in several of his stories. However, what is original – or at least, less familiar – is the genuine interest in relationships. You can see why Gyllenhaal would have thought that Jones could direct the script after he had made Moon since the mixture of SF, limited ‘action’ sequences and relationship dilemmas in that film is repeated here. I think we really do care by the end of the film what happens to each of the characters and how Stevens’ decisions affect them.

Another genre repertoire that is important is the ‘puzzle film’. At first I thought we were going to get a version of the Rashomon narrative – perhaps as in the Lola rennt (Run, Lola, Run) mode, but I think that Source Code is slightly different. Those narratives require a repeated presentation of the same events from the viewpoint of different characters (i.e. potential ‘unreliable narrators’). In Source Code, the events change each time since Stevens learns a little more about what is going on and attempts to change what happens in the eight minutes. But he doesn’t do this by ‘time travel’ so much as gaining intelligence from the past that could help to change the future. The narrative has a final twist that suggests that he also does something else – but I won’t spoil how the narrative enigma evolves over the course of the film. I suspect that many in the audience will want to watch the film a second time to check out their own understanding – I certainly will as I’m not sure I’ve got it right.

The restricted narrative space also makes this a ‘train thriller’. The obvious connection here is to Hitchcock and North by Northwest, not only in the on board train meeting (see the still above) but also the scenes in the station washroom. It’s great to see American films returning to trains – always more interesting than car chases for me. (I went straight to the Chicago Commuter Train website to check out the system – it looks impressive.) The thriller also has a contemporary feel because of the intimation of a terrorist threat – and the assumptions that creates.

Vera Farmiga as the 'Controller' and Jeffrey Wright as her boss.


There were stories in the first couple of weeks on release that this film was too clever for its own good – that some audiences weren’t bright enough to ‘get it’. This is nonsense of course. The film has reached beyond the fans of Moon I think to pick up mainstream SF fans and others. Some haven’t enjoyed the film but most have. The box office doesn’t lie and Source Code has ‘legs’ – at least in the US and UK. The weekly drop in audience numbers has been much lower than for comparable Hollywood SF films.

Like Moon, Source Code has a limited number of main characters (five) and only a few locations. Casting was clearly important and I think it works very well. Gyllenhaal is slightly unconventional for a leading man/action figure (although it may be just because I still tend to think of him as Donnie Darko) – has he always had that slightly manic eye, horribly reminding me of Tony Blair? But what most intrigued me was that I took Michelle Monaghan to be the same age or younger than Gyllenhaal and Vera Farmiga to be the ‘older woman’. In fact Monaghan is three years older than Gyllenhaal and just three years younger than Farmiga. Clearly she acts/dresses ‘younger’ but I also wonder if the uniform and her authority makes Farmiga look/feel older?Either way it’s good to have decent roles for women in SF.

Here’s the trailer (WARNING: it gives away more of the plot than this review):