Fritz Lang is not a writer or director prone to following narrative conventions too closely. While that is admirable artistically, it can make his films relatively difficult viewing and risky. It's difficult because you can't expect Lang's films to have stories that unfold in a traditional way, so as a viewer, you have to work harder, and it's risky because the experimental nature doesn't always result in a successful finished artwork. Metropolis (1927) and Fury (1936) are two examples of Lang films with unusual approaches that work extremely well. M (1931) is an example of one that isn't quite so successful in my opinion, even though many people seem to love it. Spies is somewhere in the middle.
Like M, Spies begins with more of a thematic collage. We're immediately dropped into a fast-paced, fast-cut sequence of spies stealing important documents and killing others when expedient. We also see news of this quickly filtering through both the spy world and the official media organizations. The sequence is impressive technically, but most viewers will be searching for the characters to latch onto. Also like M, Lang doesn't let viewers off that easily. He constantly introduces new characters for at least the first 40 minutes (of the 90-minute U.S. version). A number of the characters look similar, and most do not have much accompanying exposition to help viewers ground them. Making it more difficult, inter titles (this is a silent film) where characters' names are first presented often appear between two scenes with different characters, so that it's difficult to figure out which character the inter title is supposed to apply to.
Additionally, the story is complex enough and hinges on small details to an extent where it can be difficult to follow on a first viewing. I had to watch the film twice to feel confident that I had a grasp on the plot. By the end of the first viewing, you know who the principle characters are, so on the second viewing you can focus more on them rather than the countless ancillary characters who keep appearing and disappearing.
It might be difficult to count this as a flaw. There's no reason that films should be 100% accessible on a first viewing, and in fact, if you're someone who likes to watch a film more than once, a gradual unfolding on repeated exposures can be more desirable. But it's best to be forewarned. Expect to be confused unless you keep a scorecard, so to speak, and keep hitting pause.
However, once you've sorted the film out, the basic gist turns out to be fairly simple and straightforward. An anti-government (the exact government isn't specified--it's rather left intentionally vague, except that we know it's somewhere in Europe) spy organization, headed by a man named Haghi (Rudolf Klein-Rogge), is after some important treaty. There are three copies of it, and Haghi is trying to intercept them all. Haghi's principle foe is the government's Secret Service Agency, which ends up putting a James Bond-like agent named 326, or Donald Tremaine (Willy Fritsch), on the case. Haghi sends his agent Sonia (Gerda Maurus) to gain information, and hopefully the treaty, from Tremaine. He also enlists the help of a convict whom he helped bust out of death row, Hans Morrier (Louis Ralph). Complications arise when Sonia and Donald fall in love. Haghi tries to get at another Secret Service affiliate, Doctor Masimoto (Lupu Pick) through another attractive female spy, Kitty (Lien Deyers). There is a lot of double crossing, and there are a lot of spy versus spy machinations. The film focuses on these and the difficult romance, as Haghi does the typical megalomaniacal "evil genius" thing of trying to take over the world (although just how he plans to do this with a treaty and his odd combination of public vocations remains a mystery).
Lang is often thought of as a heavily visual director. In conjunction with infamous cinematographer Fritz Arno Wagner, who lensed such masterpieces as F. W. Murnau's Nosferatu (1922), Lang doesn't disappoint on that end. There are a number of conspicuously "arty" shots, such as a complex of staircases in Haghi's bank, Cabinet of Dr. Caligari (1920)-like rooftops, or the marvelously fantastical imagery of the final clown performance (you have to see the film to understand why there's a clown in it), but Lang and Wagner are just as good with more subtle compositions.
But there are flaws, too. Even viewing the film a couple times, the overly complex details of the plot can be hard to follow and many are left as dangling threads. The overall gist of the plot is good, and Spies certainly has influenced subsequent films in the genre, but there are script problems here.
Although Lang is good at creating suspense when he wants to (for most of M, he didn't seem interested in it), and he does do effectively quite a few times here, he still directs and edits action scenes a bit awkwardly. Also, the U.S. 90-minute version is satisfactorily paced, which helps the suspense, but it is also obviously sped up. The film was shot at 16 frames per minute and IMDb lists the "original" version at almost twice as long. If Spies were slowed down to run close to 3 hours, the pacing would be off. The plot would unfold too slowly.
For me, this is a middling silent film, as well as a middling Lang film, thus earning my 7 rating, or a "C", but worth watching for serious fans of thrillers, crime and espionage flicks. As always, Lang is a good "ideas" man, and this film can make you feel as if the whole world must consist of spies, or at least people whom you can't exactly trust.
