Barely hinted at is the problem of algorithmic content selection, across the various media. For years, I’ve felt that Spotify has trapped me into a musical ghetto, just feeding me more of the same stuff I’ve “liked” in the past, so I set about to teach Spotify that I’m not the listener they think I am. The “how” of that is too complicated for a comment, but it might be an interesting avenue for you to explore. I knew it was working the other day when my wife said: you’re sure listening to a lot of weird stuff these days. Exactly!
These trends began before the algorithmic revolution, but no doubt algorithms will push them forward (not that there’s much farther to go!). I share your frustration with Spotify. Any dummy can tell me that I’ll probably like They Might Be Giants if I like the Talking Heads. Can’t the machines at least tell me something useful?
Technically the "machines" are mostly just software, and can be programmed with different inputs, contexts, preferences (which gets back to your original point).
This is actually a big reason as to why I don't use Facebook much- it shows me the same 5 friends over and over unless there's a big life event like getting married. And over the years it has decided who my close friends are, even though sometimes it's just people who post a lot!
How to find more weirdos on facebook that you could get to know in five lifetimes:
step 1.
find a "controversial" individual or group. observe or join it. start making new connections with people that are interesting via that individual/group.
If you ever get around to it, I'd love to read a write up of how you taught the Spotify algorithm to show you something different... I'm have been struggling with the same problem for a while.
Well, it’s definitely on my list of things to write about, but it’s likely a couple weeks out. Pop me an email (tompendergast@gmail.com) and I can share what I’ve been doing with you (or you can subscribe to my substack)
Glad to hear you're going to write it up. My experience is that Spotify doesn't have a clue what I like, even though it should have a ton of data from which to extrapolate.
A marvelous work. Let me add one thought / caveat, though:
It is hard to square your findings with the increasing fragmentation of taste observable *uniquely* on non-traditional specialized streaming distribution channels such as Spotify or Netflix. There is evidence that what's happening on these platforms (and on these platforms only) strongly diverges from the picture you paint in your article (see here: https://twitter.com/emollick/status/1521674942928064512?s=20&t=cdkHWrB0mBurNS0MpH1R0w)? That is: while media consumption overall exhibits a pattern of rising uniformity and oligopoly, if one focuses exclusively on streaming platforms the opposite is the case.
My own hunch is that what we are facing is best described as a sort of "audience bifurcation": while a more demotic public continues to frequent the movies and has a pretty uniform cinematic experience, their more elitist peers consume motion picture content uniquely or primarily via streaming platforms and have a much more variegated and fragmented media diet. This also creates a sort of feedback loop: if the most novelty craving part of the audience migrates away from movie theaters (and from radio and cable television), Hollywood and Co. have fewer incentives to cater to that public, thereby contributing to an even greater chasm between the world of mass media and the world of streaming. This "audience bifurcation" could also help to explain Hollywood's loss of prestige, the growing irrelevance of the Oscars (as described masterfully by Ross Douthat) etc.
Ooh, interesting insights here Michael! Extending your questioning, I wonder if the streaming audience itself may have several strands--on one end of the spectrum, those who simply use it to get exposed to what the masses want, and on the other end, those who are inveterate novelty seekers (which is i think the part you were referring to)? Certainly the streaming services has user classes or personas that they know about ...
Agreed! I originally had a section on this and took it out mainly for space and because I was surprised at how little market share streaming services captured until just a few years ago. So they may speed the trends, but all of them started much earlier.
They may speed the trends, or they may simply consolidate or perpetuate the trends. If the algorithms assume that humans will behave as your trends predict (as they should), they will perpetuate their existence, right? It’s why we need the ability to “tweak” the trends more readily--to insist that we get exposed to difference.
Great essay. Have you read any Mark Fisher? He wrote a lot about this trend, focusing on a very broad scope where huge innovations in the 60s through 80s in pop culture across many mediums hit a huge wall of stagnation starting in the 90s and increasingly intensifying. He's been proven more right in every year since his passing. His explanation is mostly economical - without reliable social support that allows people to freely pursue and experiment with art, art has become a rich person's business again, with only the most reliably profitable projects being invested in or promoted. It's as convincing as it is tragic.
I think most of his writing on this subject specifically is buried deep in the archives of his blog, k-punk. I read the physical collection of it but unfortunately I don't have it handy, so I don't remember the names of particular essays that focused on it. I'd recommend looking through it if you can find it and looking for essays with provocative names, they will likely have provocative content too :) I think his book on hauntology also covers this subject, but I haven't read that yet.
Great essay. This ties in with my thoughts about today's culture: We are living in hauntological times: a stagnant period in which the past is being plundered and it seems impossible that the future will ever arrive.
Thanks for sharing that essay of yours. I’d say I enjoyed it, though I don’t suppose “enjoy” is the right word! I just kept hearing Kurt Vonnegut in the background muttering “and so it goes.”
Adam, thank you for this fantastic analysis and more than few insights.
Couldn’t agree more with the notion that trusted entertainment franchises/sequels/brands makes for easier, less-fraught decision making.
But why? Rather than risk choosing something unknown and possibly bad, choosing something known and possibly bland is the safer course.
We now live in a world where safety is heralded above all else. Understandable as we navigate a deadly pandemic but an abundance of safety rules and restrictions is not making us feel safer. Instead they’re contributing to our culture of fear.
Seeing the future as inherently unsafe limits our freedom to explore, experiment and make our own choices. Rules and restrictions implicitly communicate to us that we’re incapable of making our own risk assessments and assuming responsibility for our own life.
The modern cult of safety is infantilizing us, ensuring we remain dependent on overbearing authority figures - be it presidents of movie studios or presidents of countries - who promise to keep us safe from what we’ve been socialized to believe is a dangerous world.
Avoiding risk and dangers and anything new (even in our entertainment choices) is avoiding life.
I agree with the analyis in the article, but also with Stefano's point. Recall William Goldman's mantra that "nobody knows anything" in Hollywood. As a result, the rule is that you don't get fired for greenlighting something that looked like it should have been a success, but you do get fired for greenlighting something risky that fails -- plus you get little credit for it.
The factors in the article definitely made that even moreso, though, especially consolidation - if you have just one layer of management above you they might take a chance (Alan Ladd at Fox was the only person willing to make Star Wars; Stanley Unwin accepted the likelihood of losing money on The Lord of the Rings) but now that they have a chain of people to justify their decisions to they're going to be a lot more conservative when it comes to risk.
The twist of course is that what is "risky" is itself conventional wisdom and often provably wrong, such as the idea that audiences won't see movies with women or non-White actors in the lead role.
Thinking on it a bit, I think another element of consolidation is specifically that the consolidators were publicly owned corporations. Not that private ownership is any kind of panacea, but when George Lucas owned Lucasfilm, he could decide to go fifteen years between Star Wars movies if he wanted to; Disney literally can't, so long as Star Wars is making them money.
I know some folks who worked in film a while back, and apparently it used to be conventional wisdom in Hollywood that pirate movies don't make money. Then Pirates of the Caribbean was a huge success, and the lesson Hollywood learned was "people want more movies about Disney rides."
Hadn't thought about the public company angle––good point!
I wonder how this plays out in books. Publishers have control over who they publish and what they promote, but how much of that can explain why Danielle Steel can write so many bestselling books?
Unfortunately, this "vibrant anarchy" you point us toward is in the realm where dwells those crap efforts by amateur ninnies. There are no signposts of "quirky and weird, but done with talent" and "quirky and weird but unwatchable dreck" to help the wandering prospector.
I think there's a definite time-investment calculation going on consciously or unconsciously in a person. Wading thru the tedious drivel online for three hours to find one marvellously odd song feels like poor time management. (Especially when an excited search only discovers the deflating news this amazing song was done eleven years ago and the artist never did anything else, ever.)
Escapism has to be a huge driving force to endless sequels. A few decades ago, books that ran beyond a trilogy were given a suspicious glance and/or derisive snort. In the modern day, it's almost impossible to find a standalone novel. A mere trilogy is practically quaint. But I guess the desire to escape a big, nasty world is nothing new. So, we circle back to your essay as to why "reading" is no longer a sufficient escape, but reading serialized stories.
I think the collective efforts of a bunch of scrying eyes can cut through that vibrant anarchy. In the early web, before serious publishers took hold of things and told us what to watch, there was a way for videos/comics/projects to rise to the top. It felt like a more “town square” method—people would post about them on forums, you’d spread them by word of mouth. A lot of early internet webcomics/videos spread and gained significant popularity this way. It’s possible for a broad network of humans to serve as a selecting algorithm and promotion tool.
For sure, foraging is hard, but much more fun with friends. I think the other thing that helps is just trying to like more stuff. The most interesting art is flawed in some way that gives you something to think about.
Loved this read- thanks for your research & writing!
One thought to add to the balance: I wonder if the social consequences are not given enough weight. More homogeneity has downsides but also means individuals are more likely to have common interests- shared culture to bond over and discuss. A common diet has more appeal when you value sharing your meals.
Thanks to Roth Douthat's tweet I found this excellent essay. I tend to agree with Douthat's Decadent Society thesis, and here, where the apparent oligopoly phenomenon reflects a a general sclerotic and repetitive devolution of Western society and culture. My belief is that the underlying psychology is that most people like to be TOLD WHAT TO THINK. This seems especially true when our best thinking, our best successes, appear to be behind us, and we've become truly sheltered and unchallenged. Politics, for instance, is a hard topic. But we don't have to struggle to be objective or curious when we can be told what to think by Fox News, Twitter, or MSNBC. These become friendly, self-reinforcing places. But they are shallow and vacuous places that have germinated into a more menacing thought police pervading our institutions and national dialogue. Not quite so dark for music, movies, books, art, etc. But the same psychology is there: 'Relieve me of the burdens of objectivity, curiosity, originality, and growth - just tell me what to like and think!"
This is a superb essay, well researched, with an astonishing thesis, and supported with beautiful graphs. I kept saying things like, "Surely not! I was in second grade when Gunsmoke came on the air and shopping for nursing homes when it went off!" I arrogantly turned to Google for support only to find that 7 of the top 10 longest-running TV shows are still on the air. So you won again. But I found your paragraph on proliferation to be a persuasive explanation of the phenomenon. How often have I opened YouTube, only to immediately click it off in bewilderment and pick up a beloved P. G. Wodehouse story for the tenth time.
I don't know where you'd get the data, but I'd love to see this same work done with ROI. So not just top grossing or selling, but greatest return and maybe greatest return as a percent of investment.
In thinking about video games in particular major studios are able to produce projects on a completely different scale than indy developers. They sell way more copies, but also spend way more money on producing and, in particular, marketing their products to make sure they align with (or, more accurately, shape) the zeitgeist. Yet the Game of the Year isn't always a huge budget title. Hades was made by Super Giant Games who have a reported staff of 20 people. Compared to Rockstar who made Red Dead Redemption at over 2,000 employees. Hades sold a million copies and Red Dead 2 about 42 million. So 100x the employees but "only" 40x the number of sales.
To me it's less that there are only so many "successful" media producers, after all the team at Super Giant made incredible money on their game in relation to their investment, and more that there is an Oligopoly on our time and our eyeballs. Smaller media producers can make it, they can do well, heck they can make the best video game of the year and yet they'll only reach 1/40th of the audience.
I think that's a good point that, even though it's now way easier to deliver content to an audience, reaching the top of the chart still requires you to work on conventional platforms. You can't produce a bestselling game, book, or movie without getting into GameStop, Barnes & Noble, and AMC, as archaic as those brands seem. I have no idea how a song gets on the charts, but I imagine the path from YouTube is way harder than the path from a major label. So Hades is competing on a whole different level where they can still rake in sales, but it's nearly impossible for them to sell as much as an RDR2.
I know literally nothing about this topic but was reading the comments, and this one made me think of Tom MacDonald - check it out - he is 100% independent (music) but had been hitting #1 on billboard hits for Rap for the past couple years and just took #1 in Pop with his first ever pop song. He's also getting radio play for the first time (he didn't arrange it) ever, his trick - his fans. It's actually pretty amazing what's he managed to do.
The recent ability to sample a work of art before choosing to buy, and to read endless reviews (which themselves are in a steep darwinian struggle to grab attention) means that the pressures for optimisation are higher than ever. Pop music is now selected for radio play based on voting from target audiences who only hear 2 second grabs of tracks (this has caused a loss of dynamic and emotional range through a single song, which has flattened the narrative nature of popular music). I can read the first few pages of almost any book on amazon, meaning that they are under enormous pressure to grab me at that stage, whereas previously I would probably buy a book based on the blurb alone (or more likely flick to a random page and see if it appealed there).
It almost feels like a kind of optimisation trap that restricts surviving/successful artworks to a narrow local environment. Experimentation away from this narrow band is dangerous, and usually punished, but it means we miss out on discovering new possibilities.
I wonder if this might not be a problem of data availability and not looking into the full range of competitors. To take an example, let me talk about about my "TV viewership".
I used to watch exclusively TV shows that were either on a broadcast channel or a streaming service (e.g Netflix and Hulu). These days however, the majority of my short to medium format video consumption is channels on Youtube, and my consumption there is probably 50% a small number of channels that I watch nearly every video they produce and the other 50% smaller channels from topics the algorithm suggests.
The traditional TV shows that I still watch have definitely decreased in variety and novelty, but this doesn't reflect a decrease in variety novelty in my video consumption, but rather a shift in where I'm getting my video consumption and what portion of it remains in the old "channel" (for lack of a better word).
A similar thing has happened to my reading habits where I read many fewer traditionally publiished novels but probably spend more time overall reading long form blog posts (like this one) for non-fiction and various web serials for fiction.
Basically, what I'm arguing is that the metric we should be concerned about is not how much money is going to who in the traditional media methods but rather how much consumption time/effort is going to who for a given media _category_.
Unfortunately, this kind of data is much harder to come by, but I think that if one could get it, it would significantly complicate the story.
I wish I had the full data on all of these markets, but it appears that it either a) doesn't exist or b) is held captive by market research companies that want big bucks for it. So it's an empirical question that I can't fully answer, but there are a few reasons to doubt that the trends would reverse if we could see it all.
First, we actually do have pretty much the full data for movies, and as I show in the post, the top 20 movies actually take up a bigger share of the market than ever. It's possible that market is weird, but it's at least one case where we don't see the trend reverse when we can see everything. And although people can and do watch lots of video content elsewhere, there is something specific and special about going to a dark room with strangers and giving your full attention to a story for two hours, and something sad about that art form being overwhelmed by retreads.
Second, these trends were all in full swing long before it got easy to access lots of alternative content on the internet. Most Americans didn't have broadband until 2007, which is also the year Netflix began streaming. YouTube launched in 2005, Spotify in 2011. All of these platforms really only took off years after that. Streaming only accounted for 14% of television viewing in 2019. New consumption channels may render the top of the charts nonrepresentative, but that's either happened within the past few years, or it hasn't happened at all yet.
And third, it's easy to underestimate how big the mainstream is, especially if you're a bit outside of it. I don't have any friends who watch any of the TV shows in the top ten, and many I've never even heard of. The same was true in 2007. This happens, I think, both because social networks tend to be pretty culturally homogenous and because some kinds of content draw lots of discussion online and others don't. People love to dissect Breaking Bad, podcast about it, even wear Heisenberg t-shirts. But it got fewer viewers per episode than Chicago Fire, a show I've literally never heard anyone say anything about. (sources: https://en.wikipedia.org/wiki/Breaking_Bad#Viewership, https://deadline.com/2022/05/chicago-fire-finale-tops-wednesday-ratings-survivor-1235032651/)
So it's definitely possible that most people's content consumption has shifted like yours, and the top of the charts simply don't pick that up, and that movies are different from all other forms of media. I'd love to have the data to answer that conclusively, but until then my bet is that the data we have won't be overturned by the data we don't have.
Interesting! Both of these make the point that stars have been replaced by franchises, which is consistent with seeking familiarity—two Marvel films are going to be much more similar to each other than two Tom Hanks films. Interested to read what you write!
Excellent article. I think as with many (if not most?) things, a lot of this comes down to education. Many people with a strong liberal arts education, coupled with fairly broad life experience, actually crave things that are completely new and different (and not derivative). And for those people (myself being one) we spend our lives in a cultural desert, forever hearing breathless reviewers tout some new song or movie as "revolutionary" when, alas, it is not. (Not even remotely.) So we'll keep foraging and digging...
Very cool! One other explanation on the demand side might be increased demand for shared experiences. As communication costs have come down, it has become easier to know what everyone (and with the internet, it really can be almost everyone) is watching/reading/listening to. Perhaps in the past a greater variety of products could each catch fire in different communities while today our social networks are more interconnected. A cool economics paper related to the desire for shared experience in moviegoing is “Something to Talk About” by Gilchrist and Sands.
I just love this sentence: "Big things like to eat, defeat, and outcompete smaller things."
And I have suspected this for some time: "Indeed, maybe cultural oligopoly is merely a transition state before we reach cultural monopoly."
And this blew my mind: "There are now 60,000 free books on Project Gutenberg, Spotify says it has 78 million songs and 4 million podcast episodes, and humanity uploads 500 hours of video to YouTube every minute." (Think of the internet pollution!) As did much of the other research bursting from this blog. WOW. For someone who left not only Ohio in 1996, but also participation in the vast majority of pop culture, this blog contained an astonishing amount of information. You conquered my steep learning curve in a snap! Thanks, birthday boy!
Barely hinted at is the problem of algorithmic content selection, across the various media. For years, I’ve felt that Spotify has trapped me into a musical ghetto, just feeding me more of the same stuff I’ve “liked” in the past, so I set about to teach Spotify that I’m not the listener they think I am. The “how” of that is too complicated for a comment, but it might be an interesting avenue for you to explore. I knew it was working the other day when my wife said: you’re sure listening to a lot of weird stuff these days. Exactly!
These trends began before the algorithmic revolution, but no doubt algorithms will push them forward (not that there’s much farther to go!). I share your frustration with Spotify. Any dummy can tell me that I’ll probably like They Might Be Giants if I like the Talking Heads. Can’t the machines at least tell me something useful?
Technically the "machines" are mostly just software, and can be programmed with different inputs, contexts, preferences (which gets back to your original point).
This is actually a big reason as to why I don't use Facebook much- it shows me the same 5 friends over and over unless there's a big life event like getting married. And over the years it has decided who my close friends are, even though sometimes it's just people who post a lot!
How to find more weirdos on facebook that you could get to know in five lifetimes:
step 1.
find a "controversial" individual or group. observe or join it. start making new connections with people that are interesting via that individual/group.
step 2.
go back to step 1, repeat
step 3: get on substack or some other alt-platform (which you already did)
If you ever get around to it, I'd love to read a write up of how you taught the Spotify algorithm to show you something different... I'm have been struggling with the same problem for a while.
Well, it’s definitely on my list of things to write about, but it’s likely a couple weeks out. Pop me an email (tompendergast@gmail.com) and I can share what I’ve been doing with you (or you can subscribe to my substack)
Glad to hear you're going to write it up. My experience is that Spotify doesn't have a clue what I like, even though it should have a ton of data from which to extrapolate.
Check it out here: https://tompendergast.substack.com/p/digging-my-way-out-of-an-algorithmic. It took a real act of will to get the ship to steer in another direction, and I’m still not totally satisfied, but you can move the algorithm, I’m convinced.
Thanks; I appreciate it, and I will check it out.
Subscribed!
Also subbed. Also looking to break the algorithm. Thanks!
Thanks Chris--the article will come out next week. To prepare, start unliking everything you’ve ever liked! :)
A marvelous work. Let me add one thought / caveat, though:
It is hard to square your findings with the increasing fragmentation of taste observable *uniquely* on non-traditional specialized streaming distribution channels such as Spotify or Netflix. There is evidence that what's happening on these platforms (and on these platforms only) strongly diverges from the picture you paint in your article (see here: https://twitter.com/emollick/status/1521674942928064512?s=20&t=cdkHWrB0mBurNS0MpH1R0w)? That is: while media consumption overall exhibits a pattern of rising uniformity and oligopoly, if one focuses exclusively on streaming platforms the opposite is the case.
My own hunch is that what we are facing is best described as a sort of "audience bifurcation": while a more demotic public continues to frequent the movies and has a pretty uniform cinematic experience, their more elitist peers consume motion picture content uniquely or primarily via streaming platforms and have a much more variegated and fragmented media diet. This also creates a sort of feedback loop: if the most novelty craving part of the audience migrates away from movie theaters (and from radio and cable television), Hollywood and Co. have fewer incentives to cater to that public, thereby contributing to an even greater chasm between the world of mass media and the world of streaming. This "audience bifurcation" could also help to explain Hollywood's loss of prestige, the growing irrelevance of the Oscars (as described masterfully by Ross Douthat) etc.
Ooh, interesting insights here Michael! Extending your questioning, I wonder if the streaming audience itself may have several strands--on one end of the spectrum, those who simply use it to get exposed to what the masses want, and on the other end, those who are inveterate novelty seekers (which is i think the part you were referring to)? Certainly the streaming services has user classes or personas that they know about ...
Agreed! I originally had a section on this and took it out mainly for space and because I was surprised at how little market share streaming services captured until just a few years ago. So they may speed the trends, but all of them started much earlier.
They may speed the trends, or they may simply consolidate or perpetuate the trends. If the algorithms assume that humans will behave as your trends predict (as they should), they will perpetuate their existence, right? It’s why we need the ability to “tweak” the trends more readily--to insist that we get exposed to difference.
Great essay. Have you read any Mark Fisher? He wrote a lot about this trend, focusing on a very broad scope where huge innovations in the 60s through 80s in pop culture across many mediums hit a huge wall of stagnation starting in the 90s and increasingly intensifying. He's been proven more right in every year since his passing. His explanation is mostly economical - without reliable social support that allows people to freely pursue and experiment with art, art has become a rich person's business again, with only the most reliably profitable projects being invested in or promoted. It's as convincing as it is tragic.
I’d love to read more! Where should I start?
I think most of his writing on this subject specifically is buried deep in the archives of his blog, k-punk. I read the physical collection of it but unfortunately I don't have it handy, so I don't remember the names of particular essays that focused on it. I'd recommend looking through it if you can find it and looking for essays with provocative names, they will likely have provocative content too :) I think his book on hauntology also covers this subject, but I haven't read that yet.
I stumbled across this Mark Fisher video (https://www.youtube.com/watch?v=aCgkLICTskQ) which talks around his hauntology book, that might be interesting for you.
Great essay. This ties in with my thoughts about today's culture: We are living in hauntological times: a stagnant period in which the past is being plundered and it seems impossible that the future will ever arrive.
Cultural Dopes:
https://lareviewofbooks.org/article/cultural-dopes/
Had I found this sooner I would have referenced some of the great facts assembled here.
Proud to be doing my part adding to the hauntology as one of those “alternative and literary” Substacks! ☺️
Thanks for sharing that essay of yours. I’d say I enjoyed it, though I don’t suppose “enjoy” is the right word! I just kept hearing Kurt Vonnegut in the background muttering “and so it goes.”
Adam and GD, I used your two essays as a jumping off point for some thinking of my own. Thanks for the fruitful exchange: https://tompendergast.substack.com/p/blame-it-on-the-black-star
Thanks.
Adam, thank you for this fantastic analysis and more than few insights.
Couldn’t agree more with the notion that trusted entertainment franchises/sequels/brands makes for easier, less-fraught decision making.
But why? Rather than risk choosing something unknown and possibly bad, choosing something known and possibly bland is the safer course.
We now live in a world where safety is heralded above all else. Understandable as we navigate a deadly pandemic but an abundance of safety rules and restrictions is not making us feel safer. Instead they’re contributing to our culture of fear.
Seeing the future as inherently unsafe limits our freedom to explore, experiment and make our own choices. Rules and restrictions implicitly communicate to us that we’re incapable of making our own risk assessments and assuming responsibility for our own life.
The modern cult of safety is infantilizing us, ensuring we remain dependent on overbearing authority figures - be it presidents of movie studios or presidents of countries - who promise to keep us safe from what we’ve been socialized to believe is a dangerous world.
Avoiding risk and dangers and anything new (even in our entertainment choices) is avoiding life.
I don’t have any data linking safer media choices to an expanding definition of “safety” and “harm”, but they sure seem to have coincided
I agree with the analyis in the article, but also with Stefano's point. Recall William Goldman's mantra that "nobody knows anything" in Hollywood. As a result, the rule is that you don't get fired for greenlighting something that looked like it should have been a success, but you do get fired for greenlighting something risky that fails -- plus you get little credit for it.
The factors in the article definitely made that even moreso, though, especially consolidation - if you have just one layer of management above you they might take a chance (Alan Ladd at Fox was the only person willing to make Star Wars; Stanley Unwin accepted the likelihood of losing money on The Lord of the Rings) but now that they have a chain of people to justify their decisions to they're going to be a lot more conservative when it comes to risk.
The twist of course is that what is "risky" is itself conventional wisdom and often provably wrong, such as the idea that audiences won't see movies with women or non-White actors in the lead role.
Thinking on it a bit, I think another element of consolidation is specifically that the consolidators were publicly owned corporations. Not that private ownership is any kind of panacea, but when George Lucas owned Lucasfilm, he could decide to go fifteen years between Star Wars movies if he wanted to; Disney literally can't, so long as Star Wars is making them money.
I know some folks who worked in film a while back, and apparently it used to be conventional wisdom in Hollywood that pirate movies don't make money. Then Pirates of the Caribbean was a huge success, and the lesson Hollywood learned was "people want more movies about Disney rides."
Hadn't thought about the public company angle––good point!
I wonder how this plays out in books. Publishers have control over who they publish and what they promote, but how much of that can explain why Danielle Steel can write so many bestselling books?
Some folks on some media outlets seem to have the sole mission of stoking fear, feeding in to Stefano Boscutti's excellently articulated points above.
Unfortunately, this "vibrant anarchy" you point us toward is in the realm where dwells those crap efforts by amateur ninnies. There are no signposts of "quirky and weird, but done with talent" and "quirky and weird but unwatchable dreck" to help the wandering prospector.
I think there's a definite time-investment calculation going on consciously or unconsciously in a person. Wading thru the tedious drivel online for three hours to find one marvellously odd song feels like poor time management. (Especially when an excited search only discovers the deflating news this amazing song was done eleven years ago and the artist never did anything else, ever.)
Escapism has to be a huge driving force to endless sequels. A few decades ago, books that ran beyond a trilogy were given a suspicious glance and/or derisive snort. In the modern day, it's almost impossible to find a standalone novel. A mere trilogy is practically quaint. But I guess the desire to escape a big, nasty world is nothing new. So, we circle back to your essay as to why "reading" is no longer a sufficient escape, but reading serialized stories.
I think the collective efforts of a bunch of scrying eyes can cut through that vibrant anarchy. In the early web, before serious publishers took hold of things and told us what to watch, there was a way for videos/comics/projects to rise to the top. It felt like a more “town square” method—people would post about them on forums, you’d spread them by word of mouth. A lot of early internet webcomics/videos spread and gained significant popularity this way. It’s possible for a broad network of humans to serve as a selecting algorithm and promotion tool.
For sure, foraging is hard, but much more fun with friends. I think the other thing that helps is just trying to like more stuff. The most interesting art is flawed in some way that gives you something to think about.
Loved this read- thanks for your research & writing!
One thought to add to the balance: I wonder if the social consequences are not given enough weight. More homogeneity has downsides but also means individuals are more likely to have common interests- shared culture to bond over and discuss. A common diet has more appeal when you value sharing your meals.
Thanks to Roth Douthat's tweet I found this excellent essay. I tend to agree with Douthat's Decadent Society thesis, and here, where the apparent oligopoly phenomenon reflects a a general sclerotic and repetitive devolution of Western society and culture. My belief is that the underlying psychology is that most people like to be TOLD WHAT TO THINK. This seems especially true when our best thinking, our best successes, appear to be behind us, and we've become truly sheltered and unchallenged. Politics, for instance, is a hard topic. But we don't have to struggle to be objective or curious when we can be told what to think by Fox News, Twitter, or MSNBC. These become friendly, self-reinforcing places. But they are shallow and vacuous places that have germinated into a more menacing thought police pervading our institutions and national dialogue. Not quite so dark for music, movies, books, art, etc. But the same psychology is there: 'Relieve me of the burdens of objectivity, curiosity, originality, and growth - just tell me what to like and think!"
I haven’t read The Decadent Society—I’ve got a feeling I might disagree, but I’ll have to read and find out!
This is a superb essay, well researched, with an astonishing thesis, and supported with beautiful graphs. I kept saying things like, "Surely not! I was in second grade when Gunsmoke came on the air and shopping for nursing homes when it went off!" I arrogantly turned to Google for support only to find that 7 of the top 10 longest-running TV shows are still on the air. So you won again. But I found your paragraph on proliferation to be a persuasive explanation of the phenomenon. How often have I opened YouTube, only to immediately click it off in bewilderment and pick up a beloved P. G. Wodehouse story for the tenth time.
Ha! The wonderful thing about Wodehouse is that reading 10 stories is the same as reading one story 10 times anyway
HaHa!
I don't know where you'd get the data, but I'd love to see this same work done with ROI. So not just top grossing or selling, but greatest return and maybe greatest return as a percent of investment.
In thinking about video games in particular major studios are able to produce projects on a completely different scale than indy developers. They sell way more copies, but also spend way more money on producing and, in particular, marketing their products to make sure they align with (or, more accurately, shape) the zeitgeist. Yet the Game of the Year isn't always a huge budget title. Hades was made by Super Giant Games who have a reported staff of 20 people. Compared to Rockstar who made Red Dead Redemption at over 2,000 employees. Hades sold a million copies and Red Dead 2 about 42 million. So 100x the employees but "only" 40x the number of sales.
To me it's less that there are only so many "successful" media producers, after all the team at Super Giant made incredible money on their game in relation to their investment, and more that there is an Oligopoly on our time and our eyeballs. Smaller media producers can make it, they can do well, heck they can make the best video game of the year and yet they'll only reach 1/40th of the audience.
I think that's a good point that, even though it's now way easier to deliver content to an audience, reaching the top of the chart still requires you to work on conventional platforms. You can't produce a bestselling game, book, or movie without getting into GameStop, Barnes & Noble, and AMC, as archaic as those brands seem. I have no idea how a song gets on the charts, but I imagine the path from YouTube is way harder than the path from a major label. So Hades is competing on a whole different level where they can still rake in sales, but it's nearly impossible for them to sell as much as an RDR2.
I know literally nothing about this topic but was reading the comments, and this one made me think of Tom MacDonald - check it out - he is 100% independent (music) but had been hitting #1 on billboard hits for Rap for the past couple years and just took #1 in Pop with his first ever pop song. He's also getting radio play for the first time (he didn't arrange it) ever, his trick - his fans. It's actually pretty amazing what's he managed to do.
The recent ability to sample a work of art before choosing to buy, and to read endless reviews (which themselves are in a steep darwinian struggle to grab attention) means that the pressures for optimisation are higher than ever. Pop music is now selected for radio play based on voting from target audiences who only hear 2 second grabs of tracks (this has caused a loss of dynamic and emotional range through a single song, which has flattened the narrative nature of popular music). I can read the first few pages of almost any book on amazon, meaning that they are under enormous pressure to grab me at that stage, whereas previously I would probably buy a book based on the blurb alone (or more likely flick to a random page and see if it appealed there).
It almost feels like a kind of optimisation trap that restricts surviving/successful artworks to a narrow local environment. Experimentation away from this narrow band is dangerous, and usually punished, but it means we miss out on discovering new possibilities.
Very true
I wonder if this might not be a problem of data availability and not looking into the full range of competitors. To take an example, let me talk about about my "TV viewership".
I used to watch exclusively TV shows that were either on a broadcast channel or a streaming service (e.g Netflix and Hulu). These days however, the majority of my short to medium format video consumption is channels on Youtube, and my consumption there is probably 50% a small number of channels that I watch nearly every video they produce and the other 50% smaller channels from topics the algorithm suggests.
The traditional TV shows that I still watch have definitely decreased in variety and novelty, but this doesn't reflect a decrease in variety novelty in my video consumption, but rather a shift in where I'm getting my video consumption and what portion of it remains in the old "channel" (for lack of a better word).
A similar thing has happened to my reading habits where I read many fewer traditionally publiished novels but probably spend more time overall reading long form blog posts (like this one) for non-fiction and various web serials for fiction.
Basically, what I'm arguing is that the metric we should be concerned about is not how much money is going to who in the traditional media methods but rather how much consumption time/effort is going to who for a given media _category_.
Unfortunately, this kind of data is much harder to come by, but I think that if one could get it, it would significantly complicate the story.
I wish I had the full data on all of these markets, but it appears that it either a) doesn't exist or b) is held captive by market research companies that want big bucks for it. So it's an empirical question that I can't fully answer, but there are a few reasons to doubt that the trends would reverse if we could see it all.
First, we actually do have pretty much the full data for movies, and as I show in the post, the top 20 movies actually take up a bigger share of the market than ever. It's possible that market is weird, but it's at least one case where we don't see the trend reverse when we can see everything. And although people can and do watch lots of video content elsewhere, there is something specific and special about going to a dark room with strangers and giving your full attention to a story for two hours, and something sad about that art form being overwhelmed by retreads.
Second, these trends were all in full swing long before it got easy to access lots of alternative content on the internet. Most Americans didn't have broadband until 2007, which is also the year Netflix began streaming. YouTube launched in 2005, Spotify in 2011. All of these platforms really only took off years after that. Streaming only accounted for 14% of television viewing in 2019. New consumption channels may render the top of the charts nonrepresentative, but that's either happened within the past few years, or it hasn't happened at all yet.
And third, it's easy to underestimate how big the mainstream is, especially if you're a bit outside of it. I don't have any friends who watch any of the TV shows in the top ten, and many I've never even heard of. The same was true in 2007. This happens, I think, both because social networks tend to be pretty culturally homogenous and because some kinds of content draw lots of discussion online and others don't. People love to dissect Breaking Bad, podcast about it, even wear Heisenberg t-shirts. But it got fewer viewers per episode than Chicago Fire, a show I've literally never heard anyone say anything about. (sources: https://en.wikipedia.org/wiki/Breaking_Bad#Viewership, https://deadline.com/2022/05/chicago-fire-finale-tops-wednesday-ratings-survivor-1235032651/)
So it's definitely possible that most people's content consumption has shifted like yours, and the top of the charts simply don't pick that up, and that movies are different from all other forms of media. I'd love to have the data to answer that conclusively, but until then my bet is that the data we have won't be overturned by the data we don't have.
I have a lot of thoughts on this and will try to write something more substantive later, but I wanted to point out that there's an irony to your "Tom Hanks movie" example -- the conventional wisdom for several years at least has been that "star power" has been declining in significance, at least in the US and other English-language film markets. See, e.g., https://www.bbc.com/news/av/entertainment-arts-46813717 and https://observer.com/2018/02/will-smith-tom-cruise-jennifer-lawrence-box-office-movie-stars-are-dead/
Interesting! Both of these make the point that stars have been replaced by franchises, which is consistent with seeking familiarity—two Marvel films are going to be much more similar to each other than two Tom Hanks films. Interested to read what you write!
Excellent article. I think as with many (if not most?) things, a lot of this comes down to education. Many people with a strong liberal arts education, coupled with fairly broad life experience, actually crave things that are completely new and different (and not derivative). And for those people (myself being one) we spend our lives in a cultural desert, forever hearing breathless reviewers tout some new song or movie as "revolutionary" when, alas, it is not. (Not even remotely.) So we'll keep foraging and digging...
Really, just those of you with a liberal arts education- I call BS on this.
You're the same person who supports Tom Macdonald. GTFO here baba Canuck these nuts.
Very cool! One other explanation on the demand side might be increased demand for shared experiences. As communication costs have come down, it has become easier to know what everyone (and with the internet, it really can be almost everyone) is watching/reading/listening to. Perhaps in the past a greater variety of products could each catch fire in different communities while today our social networks are more interconnected. A cool economics paper related to the desire for shared experience in moviegoing is “Something to Talk About” by Gilchrist and Sands.
Thanks for the paper! Agreed, and rating aggregators like RottenTomatoes and Metacritic drive people to things that are already popular
I just love this sentence: "Big things like to eat, defeat, and outcompete smaller things."
And I have suspected this for some time: "Indeed, maybe cultural oligopoly is merely a transition state before we reach cultural monopoly."
And this blew my mind: "There are now 60,000 free books on Project Gutenberg, Spotify says it has 78 million songs and 4 million podcast episodes, and humanity uploads 500 hours of video to YouTube every minute." (Think of the internet pollution!) As did much of the other research bursting from this blog. WOW. For someone who left not only Ohio in 1996, but also participation in the vast majority of pop culture, this blog contained an astonishing amount of information. You conquered my steep learning curve in a snap! Thanks, birthday boy!