Bullitt (1968, Peter Yates)

Bullitt is from the period when Hollywood wasn’t calling the Mafia the Mafia yet—it’s “The Organization” here—and none of the mobsters had Italian names, but they are mostly Italian (heritage) actors. It’s especially funny because part of Bullitt’s conceit hangs on WASPs like up-and-coming senator Robert Vaughn not being able to tell Italians apart.

But that inability figures into Bullitt’s solution, which is beside the point. It’s such a nothing burger, the whole thing gets explained in two and a half lines as lead Steve McQueen and sidekick Don Gordon head off to the next set piece. Because while the film’s all about McQueen’s investigation, it’s about McQueen investigating. The film’s a character study of a hotshot San Francisco detective during one of his cases, and, despite the property damage, it might not even be one of his biggest cases. We don’t know. Vaughn wants him on the case because McQueen makes good press, but there’s never any press in the movie.

And we do see the occasional newspaper. Director Yates is hyper-focused on McQueen, though that focus doesn’t mean we get the full procedural. We don’t even see the resolution to the elaborate, exquisite car chase. Instead, we skip ahead to the next time McQueen’s going to do something idiosyncratic.

So, despite being (apparently) beloved by his fellow coppers, McQueen is very much not a regular cop. He hangs out with a happening crowd, dating British architect (I mean, she’s working on an architecture project) Jacqueline Bisset. She doesn’t know about his work life, and he likes to keep it that way. For good reason, it turns out. While there are probably a couple significant events in McQueen’s character’s work life covered in the film, the tack-on subplot about his girlfriend realizing he’s around poor people in poor places all day and not liking it seems the most consequential one.

Though, who knows, because the most relationship-building the film does for McQueen and Bisset has him being charming and then admiring. Otherwise, he’s a little busy with work.

The film opens with a gorgeous titles sequence (from Pablo Ferro Films) and expressive Lalo Schifrin music recounting a mob accountant getting away from goons in Chicago. In some ways, the titles set the tone for the film; in other ways, very much not. For instance, Schifrin’s score will barely figure in during the main action; Yates is far more interested in the diegetic sound; John K. Kean has the sound credit, with Duane Hansel, the uncredited sound editor. They do singular work. Bullitt’s got its share of genre and style innovations, but the sound design is on a whole other level.

However, the camerawork in the titles is similar to the rest of the film. Yates and cinematographer William A. Fraker alternate between vérité and precise movement. Yates likes his crane shots too, even limited ones indoors—lots of Bullitt is about watching people work and listening to the environment around them. More specifically, it’s about watching McQueen watch people work. The first major dramatic sequence in the second act involves ER doctor Georg Stanford Brown operating while McQueen (and eventually Vaughn) wait. Vaughn’s agitated, McQueen’s… seemingly not, seemingly reserved, but what’s under the surface? Yates points the camera at McQueen and inspects, which he’s already established as a motif in the first act (when McQueen’s admiring Bisset). Such good direction.

Until the third act, when Bullitt becomes a detached action thriller—with Yates, Fraker, editor Frank P. Keller, and the sound department all using previously established techniques on a giant set piece set at the airport—it’s all about watching McQueen’s face, his eyes, his breaths; waiting for him to act and react.

Other characters get similar inspection too. Usually, when dealing with McQueen or Vaughn, but also not: cabbie Robert Duvall takes it all in from the first scene in San Francisco, as the mob accountant turned witness works his way around town. Police captains Simon Oakland and Norman Fell both especially get to stare daggers while waiting on McQueen and Vaughn. But bad guys John Aprea and Bill Hickman watching McQueen (or, more accurately, his car) is maybe the most remarkable since Yates and Keller are implementing the technique in the middle of a car chase. Again, such good direction.

Most of the performances are outstanding. McQueen’s spell-binding; Oakland, Duvall, and Brown all have great moments. Vaughn’s a piece of shit politician, so he’s somewhat limited, but he’s real good at it. Similarly, Bisset’s a little too thin, but she’s fine. No time for love or architecture in Bullitt. Gordon’s a good sidekick and the occasional comic relief. He and McQueen have fantastic rapport, which makes their scenes work more than the dialogue.

The script—credited to Alan Trustman and Harry Kleiner, based on a Robert L. Fish’s novel not starring a character named Bullitt (and written under the pen name Robert L. Pike)—is terse and willfully obtuse at times. Bullitt feels like Yates and Keller, especially, made it in the editing studio, but who knows, maybe Trustman and Kleiner really did write it so remote. There are some great one-liners, though; it’s not overtly macho but enthusiastic about its procedural jargon–such a strange, transfixing combination.

Fraker’s photography is glorious and would be the easy technical standout if it weren’t for Keller’s peerless cutting.

The third act’s got a handful of problems, but Bullitt weathers them well thanks to McQueen, Yates, Keller, Fraker, and company. It’s a masterful piece of work.

Colossus: The Forbin Project (1970, Joseph Sargent)

Colossus is a pre-disaster movie, in the Irwin Allen sense. It has a lot in common with films like The Andromeda Strain and The Satan Bug. The problem is established and then the film’s story is an attempt to resolve it. It’s a little less character-oriented than the Allen disaster formula–Colosuss doesn’t have much lead-in, it sort of just gets started after the opening sequence–but that lack of character development makes the cast all the more important. The viewer’s never going to have a chance to get to know these people; the casting has to be superb.

And Colossus is perfectly cast. Eric Braeden–who I just discovered ended up on a soap (and I even recognize him)–turns in a likable, funny leading man performance. He’s always believable as the world’s foremost computer designer, but he still can get away with being a traditional (and excellent) leading man. Susan Clark’s second-billed and sort of around for half the movie in the background before she gets to take a more central role and she’s got some fantastic moments. I figured–based on that Planet of the Apes movie he was in–Braeden would be good. So the real surprise is Gordon Pinsent as the President. Too often, movie presidents aren’t convincing–or they’re played by big name actors who assume their recognizable name will make them a good president-in-crisis. Pinsent does have a lot of good material–he’s second lead for the first half–but his performance is rather impressive.

Colossus is a from-the-top crisis story. We don’t really get to see how regular people are reacting, which has become the norm today. Everyone in the film has been on the phone with the President of the United States. What director Sargent and screenwriter James Bridges have to do is make a film without special effects–we don’t see any of the disasters–work from a couple rooms. There’s the White House and there’s Braeden’s computer lab. The film could practically work as a play.

Sargent’s widescreen composition is peculiar and effective. He started on TV and he tends to use the Panavision frame to horizontally expand what would otherwise be television composition. The result is unexpected. It’s like Sargent’s composition ends up looking like deliberate, thoughtful art, when it appears to just be a pragmatic approach to widescreen filmmaking.

Bridges’s script is competent and unambitious. Colossus is from a novel–which probably followed most of the same story beats–so all Bridges has to do is make it play right. And, given how the beats develop, it’d be impossible not to. There’s some character development, left nicely with Braeden and Clark, but a lot of the script is just perfunctory. That mechanical approach ends up hurting Colossus, because there’s no sense of anything escalating. Eventually, the movie just stops. I figured there were another fifteen minutes, but no, it was end credit time.

Perceiving the passage of time in the story is partially Bridges’s fault–three days pass without acknowledgment, a problem in a story set over a specific period–but a lot of it lies on Michel Colombier’s score. It’s anti-climatic and rote. Colombier tries for melodrama and ends up wasting a lot of time.

Colossus is a boring and intriguing film. Even with the narrative distance, the characters’ dilemmas are compelling. And, at just after the halfway point, when the computer taking over the world starts talking, it gets real funny. Not dumb funny, smart funny. But still real funny. It sort of suggests the film could have cut fifteen minutes and run another thirty and it would have turned out better.