What Can Tomatoes Tell Us About Popular Music?
More than you might think!
I don’t remember precisely when heirloom tomatoes started to become a thing. But suddenly, there they were, all different colors, sizes, and shapes. They cost a little more than regular tomatoes and were only sporadically available, and at first, I didn’t pay much attention.
But many people I knew who were proud of their cooking began paying extra to get heirloom tomatoes. Restaurants made sure to note in their menus that their recipes used heirloom tomatoes, which tipped me off that this was something worth paying attention to.
There was a good reason for all this interest: they tasted great. In addition to all the different colors, sizes, and shapes, they came in various flavors. Some were more tart, others more savory. Some had overtones of other fruits or vegetables. Most were sweeter than regular tomatoes, and all were vastly more “tomato-y.”
Somewhere along the line, I realized our terminology is all wrong. “Heirloom” implies something old and rare, like an antique. “Regular” implies “normal”. Only, it’s the heirloom tomatoes that are normal. They are the remaining survivors of the hundreds of varieties of tomatoes that once existed in the natural world and strains from the parts of South America where they were first domesticated.
Meanwhile, the store-bought “normal” tomatoes are the result of centuries of selective breeding and, more recently, genetic engineering aimed at making a fruit that looks uniform, ships well, and doesn’t go bad. These are the bright red, tasteless orbs you can buy off the shelf at Safeway. They are literally clones, franken-tomatoes that have taken over the world. It is they that are the actual freaks.
Back in the 1990s, I knew the guy who founded the first company to commercialize a genetically engineered tomato. His company used an already tasteless hybrid strain as its starting point and added genes to ensure it would practically never go bad. They held a launch party to celebrate the first commercial sales of these “tomatoes,” and at the reception, they offered us samples.
After duly sampling one, I turned to my friend and said, “I hate to break it to you, but it doesn’t taste like anything.” He beamed at me and said, “Sure, but they keep two weeks in a shipping crate.”
I replied, “But I like tomatoes that taste like tomatoes.” His reply has stuck with me ever since.
“Me too, but look, we’re old farts. When we’re gone, no one will care how tomatoes used to taste.” He was grinning. I saw no hint of irony or regret.
However, many farmers and scientists became alarmed at what was happening. Some of them made sure to preserve old strains of tomatoes, along with other plants in danger of genocide—corn, beans, squash, bananas, and more.
This is not simply a matter of indulgence. We now know that monoculture is an all-or-nothing proposition. Genetic diversity is essential to the future of these crops — without it, one plague could bring starvation to millions.
Thank goodness for those who had the foresight to collect the seeds of threatened strains. Those repositories are the only places where most of them survive.
What does this have to do with music? Well, I’ll get right to the point: American popular music has been going through something like what happened to tomatoes.
Of course, there is a cottage industry of people of a certain age complaining that today’s music isn’t as good as what they grew up listening to. Some of this is generational chauvinism. I get that — it’s nothing new! For example, according to Plato, Socrates often ranted about the shallowness and trashiness of the youth culture of his day.
To this charge, however, I plead not guilty for the simple reason that I have no particular loyalty to the popular music of my generation. The generation just before mine? They had the good stuff. They gloried in the Beatles, Cream, Hendrix, the Doors, Janis Joplin, Jefferson Airplane, etc. My generation booted all of those artists off the charts in favor of Barry Manilow, John Denver, and Olivia Newton-John.
Remember, this was before the internet, so radio was the main way we heard new songs. To get to something good, we were forced to endure abominations like “Raindrops Keep Falling,” “Billy, Don’t Be a Hero,” and “Muskrat Love.”
Eventually, punk rock and New Wave arrived, like a life preserver thrown to us at the last second. My point is that generational chauvinism is not the issue here. I freely admit our generation did not have the best music. The generations just before and after ours fared better.
Most of the popular music (including rock) I listen to these days comes from young artists, people who were not around five or ten years ago. That’s what I search for. New music by new artists: That’s the main point of this newsletter. And that is where the tomatoes come in.
Unfortunately, in my search for new music, I am finding more and more sameness. Picking out the gems takes a lot of sorting through, well, supermarket tomatoes.
With apologies to Queen: Is this real? Or is it just fantasy? If it’s real, how did it happen?
Well, it turns out these aren’t just the rantings of an old curmudgeon. Data analysis shows quantitatively that this standardization process is real, progressive, and can be measured. Hit songs have become simpler and more musically repetitive.
Here is some of the evidence:
https://english.elpais.com/science-tech/2024-07-05/the-biggest-hit-songs-have-increasingly-simple-and-repetitive-melodies.html?utm_source=pocket-newtab-en-us
It has proceeded inexorably for more than 60 years, predating the internet. The era of online content and streaming is just turbocharging the trend.
By definition, the algorithms are designed to show us music similar to what we already like. Thus, by definition, they are biased against experimentation by the artist and the listener. How can this have any outcome other than reducing overall music variety and complexity?
Before going any further, let me be clear: None of this should be taken as a value judgment against simplicity itself. “Different Drum,” “Hey Jude,” and “Somebody to Love” are three of the greatest pop or rock songs of all time. They are all strikingly simple, with limited chord progressions and simple melodies built around basic scales.
The catch is that, as with all great art, once it’s been done, there is diminishing value in someone else doing it again.
Unfortunately, the music industry sees it the opposite way: If it’s been done once, let’s keep doing it as long as we can keep selling it. It’s gotten to the point where songwriters are constantly suing each other over copyright violations. The people doing the suing usually lose because so many songs use the same progressions that just sounding the same is no longer considered presumptive evidence of copying.
Daniel Parris, author of @Stat Significant, a column devoted to statistical analysis of media and the arts, has done some impressive studies of song construction. In one paper, he examined patterns in music and how they correlate with popularity.
He leads with one of his more counterintuitive findings: streaming, contrary to what its detractors claim, has democratized access to audiences in that more artists make the charts than ever before.
So what’s the catch? There is a considerable downside: while more people are charting, they sound more and more alike. There may be more of them, but there is less variety in how they write music. Democratization means anyone can make it, but only if they do what everyone else does. It sure looks like confirmation of what I thought I was hearing.
How Has Music Changed Since the 1950s? A Statistical Analysis
If you accept that the trend toward decreasing variety and complexity in music is real, the second question is easier to answer: How did it happen?
You can argue that it started in the 1950s with a concentration on the radio broadcasting industry and the growing pressure for “radio-friendly” songs. But it really took off in the 1970s, with the rise of commercial songwriting factories, consisting of offices full of trained songwriters cranking out songs for mass radio play and the “45” format. These songwriting sweatshops (they are still around) work from scripted prompts based on what has worked before.
The next significant development was the emergence of a handful of super-song-writers in the 1990s. Interestingly, several of them were from Sweden, and my guess is ABBA must have influenced them.
ABBA were revolutionary in its time. They were such a hit factory that people studied them to figure out how they did it. Those who figured out the secret could do very well for themselves. A few of them became an informal “club,” an oligarchy of top songwriters.
Though they deny it, these songwriters usually start with formulas for the simple reason that the formulas work. This small “club” of songwriters took over a huge share of the Western popular music market. How huge? Eight songwriters have written or co-written a quarter of all the songs making the Billboard pop charts over the past quarter century. Take a moment to process that.
A lot of the rest has been written by committees of journeyman songwriters working on salary or commission, using techniques taught by guys like Asaf Peres:
Learn the techniques that TOP Songwriters use & elevate your songwriting to the next level.
For a deeper look at who these people are and how they do what they do, check out my column on “The Songwriters Who Ate America”:
Peres denies he is teaching formulas, but the heart of his toolbox is something called “melodic math”, which includes, among other things, step by step instructions for getting certain effects (sad, happy, etc.) He explains exactly how to write a “hook,” among other things. You are free to call this whatever you wish.
Meanwhile, the major labels are increasingly risk averse. They prefer to publish stuff that sounds like something already successful. In other words, they prefer formulaic songs. Here is one hilarious take on the result:
Decades of industrial engineering have systematically sanded away the edges that used to give music its variety and unpredictability. This problem is with the music industry's business model, not with the artists themselves. It also seems like a pretty clear answer to my second question above.
But it gets more problematic.
More recently, computer algorithms were brought in to assist with writing hooks. We know that formulas work for making catchy hooks and riffs, and it was just a matter of time before people figured out how to program computers to do this. Now, AI is being groomed to take over the complete crafting of songs.
The fundamental limitation of AI so far is that it has to be “trained.” Training requires the processing and integration of huge amounts of data.
If you want to teach an AI to write Haiku, you feed it every example of Haiku you can find. AIs are good at writing passable Haiku after sufficient training. There are dozens of free Haiku generators available online.
Here is one — try it out for laughs:
https://boredhumans.com/haiku.php
The inherent limitation of this technology is that everything they produce is based on the examples used in their training.
No difference for music. So far, the result of using AI to write songs is what one might expect: reviewers find the songs derivative and cliché. But what if the AIs get the hang of writing hooks that hook people? You don’t need me to tell you where this is headed.
I have heard some people argue, "As long as people like it, what does it matter how music was produced?” I would reply that that ignores the human, person-to-person connection that was a part of music from its earliest beginnings. Is placing a value on that interpersonal connection merely a matter of personal taste, or, worse, simply naïve sentimentality?
Some people say it is the latter. So far, personal connection has been essential to lasting success for musicians, but it is possible that social norms will change. Mozart is indeed long dead — we don’t love his music because we have personally seen one of his recitals. If AI can compose at that level, perhaps all bets are off.
However, even if AI music becomes good enough that most people love it, there will likely be a devoted “club” of fans who seek out innovative, hand-made music.
The best comparison may be chess. By 2000, specialized computers could beat the grandmasters. By 2005, off-the-shelf software you could run on your laptop would routinely beat any human player.
https://www.smithsonianmag.com/smart-news/what-first-man-lose-computer-said-about-chess-21st-century-180962046
Despite this, very few people other than chess theorists watch computers play chess. Human chess championships are still major news stories, and online streaming and chess podcasts are exploding in popularity. The drama of watching real people match wits or try to push their own limits remains compelling for many of us.
This suggests that while the flattening of popular music will continue, it will not stamp out human creativity, originality, the drive to make great art, or the audience’s human need to connect with the artist. That doesn’t mean we should be complacent. I believe it is important for us to push back against that flattening.
How? First, it requires acknowledgment that the homogenization of music, while perhaps not an outright conspiracy, is real and is impoverishing our auditory lives. Acknowledging that means being aware of the phenomenon in the first place and resisting it. The push-back comes when we reject formulaic songwriting and actively embrace the search for music that defies our expectations, music that isn’t “easy.”
Here is one example of something that isn’t easy, but I can listen to over and over because on every listening, I hear new things:
Polyphia “Goose”
Addendum
One elephant in the room needs to be addressed, and it merits a separate article. Sometimes, there are cultural discontinuities that can profoundly affect the arts. They can happen for many reasons (war, genocide, technological change, migration, etc.), but an innovation that simply goes viral is one of them.
Specifically, the “El Pais” article cited above mentioned the advent of hip-hop as a key event driving change in musical writing. Hip-hop represents a drastic shift from the primacy of melody and harmony in what we call “music” to the primacy of lyrics.
By the way, if you are looking for a clear example of AI’s limitations, this is a good one. It is hard to imagine AI inventing something like hip-hop after being trained in traditional pop music.
Hip-hop is now the dominant popular music genre in the U.S. and is taking over a big piece of the rest of the world. It has not only changed how music sounds but also what kinds of jobs there are in music writing and production. One of the biggest effects has been reducing the importance of the “musical” component of music (see the Parris article cited above).
Is this good or bad? I say neither. Early human music, based on the evidence, started with simple vocal chanting and percussive accompaniment (drumming). All of the formal instrumentation, melody, and composition theory we call “music” is a relatively recent innovation. Viewed this way, hip-hop is retro!
It appears to me that popular music is undergoing a fission into tone-based and lyric-based art forms. The criteria for evaluating them and the reasons for loving them are becoming separate. I would love to hear from our hip-hop experts what you think.
Some traditionalists rail against hip-hop because it isn’t “real music,” but this may become as relevant as railing against motorcycles because they don’t have sails or rudders. It is a category error. And, of course, there will continue to be artists who bridge the gap. The evolution is likely still in its early stages.
Postscript
I recently made a point to check for heirloom tomatoes in three supermarkets I frequent and our local produce market. Two of the supermarkets and the produce market carried at least some heirlooms. The quantities were tiny compared with the large bins piled high with uniform, bright-red, nearly unbreakable franken-tomatoes, but they were there.
There is still a market for heirloom tomatoes, with their strange shapes, odd colors, and, most of all, their amazing and different flavors. They aren’t even that much more expensive anymore, which is a sign that they are not going away.
The analogy between music and tomatoes breaks down in the trade-offs. Franken-crops have one major advantage: they enable us to feed huge numbers of people at an affordable price. Get rid of them, and there would be widespread famine.
There is no parallel benefit to standardized, mass-produced art. Art is the only thing that sets humans apart from, for instance, termites. Termites do almost all the things we do (build cities, make war, farm, enslave other tribes, etc.). The exception, as far as we know, is art.
Termites don’t make art, and it has no immediate survival value. We make it because we appreciate and love beauty. Reducing it to “stuff” erases its entire purpose.
The one benefit of the current commercial reality for music is that unusual things can be had for a bargain. The internet has given us access to the most obscure corners of the musical universe for free. For all their efforts (and those are ongoing), the big labels and promoters no longer control what we can choose to listen to. We are still free to choose.
That freedom can only be a good thing.
Notes
The “flattening” of music can be seen as part of a generalized flattening of our broader culture. Here is an interesting perspective on that idea:
Matthew in Free Factor:
Western Culture Has Expired
Here are the articles referenced earlier in this piece:
The biggest hit songs have increasingly simple and repetitive melodies
Daniel Parris, “Stat Significant”:
How Has Music Changed Since the 1950s? A Statistical Analysis
Asaf Peres:
Learn the techniques that TOP Songwriters use & elevate your songwriting to the next level.
“The Songwriters Who Ate America”:
One other factor in the flattening of what gets played on radio was the shift from individual local DJs and program directors having some flexibility to add a song they believed in to the days when entire networks were programmed by consultants. When you look at a lot of the quirkier hits of the 70s and 80s, they gained traction because some local station played it and it grew from there.
On the other hand, the "DJ believed in it" method was also responsible for the Neil Diamond/Barbra Streisand hit "You Don't Bring Me Flowers,' so it wasn't a perfect system.
I went out after reading this and bought some heirloom tomatoes!
I'm quoting you in my next post. So many great points here.