Perfect Days is a beautiful, lyrical slice-of-life story from veteran director Wim Wenders, making his first film in Japanese, with a superb performance from K?ji Yakusho as a toilet cleaner in Tokyo who seems to find happiness in the simplicity of his daily routine. It earned Wenders his best reviews since his signature film, Wings of Desire, came out in 1987. I just wish it wasn’t so monotonous and inert, even with such a fantastic lead. (You can rent it on iTunes, Amazon, etc.)
Yakusho plays Hirayama, who cleans public toilets in a fancy neighborhood of Tokyo and lives a spartan life built around reading, eating, and listening to music. He’s a solitary person and seems to want it that way, barely talking to anyone through his daily route – especially not his incredibly annoying co-worker, Takashi – and visiting the same few restaurants and the same used bookstore and the same park to eat lunch, and while he’s driving he listens to the same small set of cassette tapes of music from the 1960s and 1970s. He takes tremendous pride in his job, using a tiny mirror like a dentist’s to make sure the undersides of fixtures are clean, and appears to have his route and work timed to the minute. His routine is interrupted a few times throughout the movie – his whiny, arrested-development coworker Takashi, who barely cleans anything, cadges money and a ride off him; his teenaged niece shows up, having run away from home – but he’s mostly stoic throughout. That is, he’s stoic until two encounters shake him enough to get him to show some real emotion: a visit from his sister, whose appearance makes it clear that Hirayama has chosen to live this somewhat ascetic existence; and an incident where he sees the restauarant owner who seems to flirt with him whenever he comes in hugging another man, which leads to a very surprising meeting that I thought was the film’s strongest scene.
In many ways, Perfect Days should be right up my alley: It’s small in scope and story, with a modest character list, and the emotions it generates in the viewer are real and well-earned. The script has a ton of heart and respects its protagonist. But after seeing Hirayama get up and go through his morning routine for the fifth or sixth time, my attention started flagging. The film may very well be asking you to ask whether this is a man who’s found happiness in a simpler existence or whether there’s something pathetic about someone who has chosen to partake so little in the modern world or enjoy the company of others. If so, it doesn’t push hard enough in that direction, even with the two scenes at the end that should at least give the script a chance to explore more of Hirayama’s character; instead, all we get is seeing him cry, the first time he shows any real emotions other than annoyance or mild pleasure in the entire film.
The film has few side characters, and the one with the most screen time, Takashi, is the most annoying character I saw in any movie other than maybe May December. He’s ridiculous, but not in a funny way. He exists just to give Hirayama something more to do than eat, sleep, and read, but he wears out his welcome before his first scene is over – and then he comes back multiple times. Hirayama’s niece has the opposite problem – she’s almost a cipher, with very little personality of her own. There’s the hint that perhaps she’s more like her uncle than she is like her own mother, but the film doesn’t explore that angle before she returns home.
Perfect Days does have a great soundtrack, comprising mostly the songs that Hirayama listens to in his van, with tracks from The Animals, The Velvet Underground, Patti Smith, The Kinks, and Nina Simone. There’s nothing in the film from later than about 1979, so we can infer that Hirayama has no interest in newer music and prefers the music of his youth – perhaps feeling that those songs are enough for him, or perhaps because he just has no interest in anything more modern. There are ideas in here, certainly, but the script doesn’t show the curiosity to learn more about its main character. Takusho’s strong turn is largely wasted here in a film that looks beautiful but never fully engages with its subject. I had high expectations for Perfect Days, but in the end, it just couldn’t hold my attention all the way through.
Filterworld.
In his new book Filterworld: How Algorithms Flattened Culture, journalist Kyle Chayka details the myriad ways in which we are thrust towards homogeneity in music, television, movies, books, and even architecture and travel because, in his view, of the tyranny of the algorithm. The book is more of a polemic than a work of research, filled with personal anecdotes and quotes from philosophers as well as observers of culture, and while Chayka is somewhat correct in that a small number of companies are now determining what people watch, listen to, and read, that’s always been true – it’s just happening now by algorithm when technology was supposed to democratize access to culture.
Chayka’s premise is sound on its surface: Major tech companies now depend on maintaining your attention to hold or increase revenues, and they do that via algorithm. Netflix’s algorithm keeps recommending movies and shows it believes you’ll watch – not that you will like, but that you will watch, or at least not turn off – thus keeping you as a customer. Spotify’s auto-generated playlists largely serve you artists and songs that are similar to ones you’ve already liked, or at least have already listened to, as I’ve learned recently because I listened to one song by the rapper Werdperfect that a friend sent me and now Spotify puts Werdperfect on every god damned playlist it makes for me. Facebook, Twitter, Instagram, Tiktok, and their ilk all use algorithms to show you what will keep you engaged, not what you asked to see via your following list. Amazon’s recommendations are more straightforward, giving you products its algorithm thinks you’ll buy based on other things you’ve bought.
Chayka goes one further, though, arguing that algorithmic tyranny extends into meatspace, using it to explain the ubiquity of Brooklyn-style coffee shops, with sparse décor, subway tiles, exposed wood, and industrial lighting. He uses it to explain homogeneity in Airbnb listings, arguing that property owners must determine what the algorithm wants and optimize their spaces to maximize their earnings. He is ultimately arguing that we will all look the same, sound the same, wear the same clothes, live in the same spaces, drink the same expensive lattes, and so on, because of the algorithms.
To this I say: No shit. It’s called capitalism, and the algorithm itself is not the disease, but a symptom.
Businesses exist to make money, and in a competitive marketplace, that’s generally a good thing – it drives innovation and forces individual companies to respond to customer demand or lose market share to competitors. These market forces led to the advent of mass production over a century ago, a process that depended on relatively uniform tastes across a broad spectrum of consumers, because mass-producing anything economically depends on that uniformity. You can’t mass-produce custom clothes, by definition. Companies that have invested heavily in capital to mass produce their widgets will then work to further expand their customer base by encouraging homogeneity in tastes – thus the push for certain fashions to be “in” this year (as they were twenty years prior), or the marketing put behind specific books or songs or movies to try to gain mass adoption. Coffee shops adopt similar looks because customers like that familiarity, for the same reason that McDonald’s became a global giant – you walk into any McDonald’s in the world and you by and large know what to expect, from how it looks to what’s on the menu. This isn’t new. In fact, the idea of the algorithm isn’t even new; it is the technology that is new, as companies can implement their algorithms at a speed and scale that was unthinkable two decades earlier.
Furthermore, we are living in a time of limited competition, closer to what our forefathers faced in the trust era than what our parents faced in the 1980s. There is no comparably-sized competitor to Amazon. Spotify dominates music streaming. Each social media entity I listed earlier has no direct competition; they compete with each other, but each serves a different need or desire from consumers. The decline of U.S. antitrust enforcement since the Reagan era has exacerbated the problem. Fewer producers will indeed produce less variety in products.
However, the same technology that Chayka decries throughout Filterworld has flattened more than culture – it has flattened the hierarchy that led to homogeneity in culture from the 1950s through the 1990s. Music was forced, kicking and screaming, to give up its bundling practice, where you could purchase only a few individual songs but otherwise had to purchase entire albums to hear specific titles, by Napster and other file-sharing software products. Now, through streaming services, not only can any artist bypass the traditional record-label gatekeepers, but would-be “curators” can find, identify, and recommend these artists and their songs, the way that only DJs at truly independent radio stations could do in earlier eras. (And yes, I hope that I am one of those curators. My monthly playlists are the product of endless exploration on my own, with a little help from the Spotify algorithm on the Release Radar playlists, but mostly just me messing around and looking for new music.) Goodreads is a hot mess, owned by Amazon and boosting the Colleen Hoovers of the world, but it’s also really easy to find people who read a lot of books and can recommend the ones they like. (Cough.) Movies, food, travel, television, and so on are all now easier to consume, and if you are overwhelmed by the number and variety of choices, it’s easier to find people who can guide you through it. I try to be that type of guide for you when it comes to music and books and board games, and to some extent to restaurants. When it comes to television, I read Alan Sepinwall. When it comes to movies, I listen to Will Leitch & Tim Grierson, and I read Christy Lemire, and I bother Chris Crawford. I also just talk to my friends and see what they like. I have book friends, movie friends, game friends, coffee friends, rum friends, and so on. The algorithms, and the companies that deploy them, don’t decide for me because I made the very easy choice to decide for myself.
So I didn’t really buy Chayka’s conclusions in Filterworld, even though I thought the premise was sound and deserved this sort of exploration. I also found the writing in the book to be dull, unfortunately, with the sort of dry quality of academic writing without the sort of rigor that you might see in a research paper. I could have lived with that if he’d sold me better on his arguments, but he gives too little attention to points that might truly matter, such as privacy regulations in the E.U. and the lack thereof in the U.S., and too much weight to algorithms that will only affect your life if you let them.
Next up: Angela Carter’s The Magic Toyshop.