If you read as many biographies and autobiographies of women artists in the rock era as I do, you come to realize how often women in that era were denied songwriting credits by their labels and producers even when they contributed significantly to a song. This is a longstanding problem in an extremely male-dominated industry where major female artists have always struggled against being told what they are allowed to sing, wear, etc.
So I don't doubt your analysis given the data that exist. The problem is with the dataset itself, which doesn't reflect the reality of who actually wrote the songs.
Excellent summation. As you point out, the rock model (self written, virtuosic) was short lived, and there are other models for commercial success. The American songbook was entirely built on professional writers providing material to expert interpreters. That was also true for Western classical music--other than Chopin, Liszt and a few others, the great classical composers were not the main performers of their work.
You have yourself published articles here pointing out that popular music is getting more uniform, or standardized, and less internally "interesting". My own opinion is that this homogenization is not a good thing. It's not just a matter of taste--brains exposed to less variety of stimuli lose some of their flexibility and even physically lose neural connections. Your thoughts on the intersection of your present article and your earlier work?
Finally, a question: what do you foresee happening with the emergent use of AI to write songs? Given that AI by definition requires training, is originality endangered?
Totally agree with you on a couple of points: brains feed and grow on complexity and “new” aka expanding consciousness, and AI song creation may be quick, but still requires knowing how to use the software, which genre to choose to match your lyrics to, when to use your own melodies, and deciding which final output song to go with. It’s not as simple as going with the first song result if you’re particular and want the best result.
Thank you! Yes, the interaction between user and AI is complex, and we are still just learning, even as the AI itself is evolving fast. Perhaps the human component will remain critical. I suspect that for mass market entertainment, there will be increasing numbers of things that are made by AI with almost no human judgement involved. People are already being fooled by short youtube memes that reasonably proficient AI users can spot right away. So the bar is low, and may get lower.
This is super interesting, Daniel! I'm curious tho - did Taylor Swift's "All Too Well" 10-minute version not meet eligibility for #1 songs written by only women? Or is it that it just has to be 1 woman to be eligible. I see that Liz Rose co-wrote it with her, just curious!
How do changing crediting practices affect this data? There are a lot of rock songs that if written today would have more credited writers, because it was common practice to just credit the *main* writer and not the producer-writer, or the musician who contributed a defining musical hook. The 60s equivalent of Max Martin may well have had a producer credit rather than a writing credit. Also, the rise of sampling culture changed the approach to crediting.
“Scan the songwriting credits for recent Taylor Swift or Sabrina Carpenter releases, and you’ll see familiar names appearing alongside the artist—including … Dr. Luke.”
What recent Swift or Carpenter song had a songwriting credit with Dr Luke? In fact, is there any “release” by either artist that had a song writing credit for him? (There’s a song by another artist that both Swift and Dr Luke had a songwriting credit for, but that’s not a Swift release).
Music authorship today is hybrid. Many chart-topping artists co-write their songs within collaborative “writer room” systems reminiscent of the Brill Building, while still maintaining personal branding. Unlike the Beatles’ rockist ideal of sole authorship, <a href="https://kendisverden.com/mathias-gidsel-alder/">hvor gammel er mathias gidsel</a> modern pop values collaboration, specialization, and production teams adapted to streaming speed, genre fluidity, and global markets.
The data is the data, but you told the story. Thank you for sharing and researching these insights. It was very enlightening. I think in general now, there really arent genres any more in the traditional sense. Due to streaming, things have gone back to the way radio was in the 60s/70s where for the most part everything was played if it was good. People are their own radio stations. Genres are now just a way to describe the sound and train the algorithms, but they dont determine the plays on streaming. They influence it.
If you read as many biographies and autobiographies of women artists in the rock era as I do, you come to realize how often women in that era were denied songwriting credits by their labels and producers even when they contributed significantly to a song. This is a longstanding problem in an extremely male-dominated industry where major female artists have always struggled against being told what they are allowed to sing, wear, etc.
So I don't doubt your analysis given the data that exist. The problem is with the dataset itself, which doesn't reflect the reality of who actually wrote the songs.
A surprisingly late example is that Cher re-wrote the verses of ‘Believe’ but did not get a writing credit.
Excellent summation. As you point out, the rock model (self written, virtuosic) was short lived, and there are other models for commercial success. The American songbook was entirely built on professional writers providing material to expert interpreters. That was also true for Western classical music--other than Chopin, Liszt and a few others, the great classical composers were not the main performers of their work.
You have yourself published articles here pointing out that popular music is getting more uniform, or standardized, and less internally "interesting". My own opinion is that this homogenization is not a good thing. It's not just a matter of taste--brains exposed to less variety of stimuli lose some of their flexibility and even physically lose neural connections. Your thoughts on the intersection of your present article and your earlier work?
Finally, a question: what do you foresee happening with the emergent use of AI to write songs? Given that AI by definition requires training, is originality endangered?
Thanks again for posting this! It is a good complement to my earlier piece on "The songwriters who ate America" https://zapatosjam.substack.com/p/the-songwriters-who-ate-america-part for those who are interested.
Totally agree with you on a couple of points: brains feed and grow on complexity and “new” aka expanding consciousness, and AI song creation may be quick, but still requires knowing how to use the software, which genre to choose to match your lyrics to, when to use your own melodies, and deciding which final output song to go with. It’s not as simple as going with the first song result if you’re particular and want the best result.
Thank you!
Thank you! Yes, the interaction between user and AI is complex, and we are still just learning, even as the AI itself is evolving fast. Perhaps the human component will remain critical. I suspect that for mass market entertainment, there will be increasing numbers of things that are made by AI with almost no human judgement involved. People are already being fooled by short youtube memes that reasonably proficient AI users can spot right away. So the bar is low, and may get lower.
This is super interesting, Daniel! I'm curious tho - did Taylor Swift's "All Too Well" 10-minute version not meet eligibility for #1 songs written by only women? Or is it that it just has to be 1 woman to be eligible. I see that Liz Rose co-wrote it with her, just curious!
My fault here. I miscoded it in the data. This would be the only recent number one hit purely written by women
How do changing crediting practices affect this data? There are a lot of rock songs that if written today would have more credited writers, because it was common practice to just credit the *main* writer and not the producer-writer, or the musician who contributed a defining musical hook. The 60s equivalent of Max Martin may well have had a producer credit rather than a writing credit. Also, the rise of sampling culture changed the approach to crediting.
Using today's crediting techniques, George Martin would get his name in there with Lennon/McCartney
Incredible write-up! Thanks for doing this, Daniel.
I was surprised to see so many non-artist songwriters on your list, and that I had never heard of them before. Great work!
“Scan the songwriting credits for recent Taylor Swift or Sabrina Carpenter releases, and you’ll see familiar names appearing alongside the artist—including … Dr. Luke.”
What recent Swift or Carpenter song had a songwriting credit with Dr Luke? In fact, is there any “release” by either artist that had a song writing credit for him? (There’s a song by another artist that both Swift and Dr Luke had a songwriting credit for, but that’s not a Swift release).
Music authorship today is hybrid. Many chart-topping artists co-write their songs within collaborative “writer room” systems reminiscent of the Brill Building, while still maintaining personal branding. Unlike the Beatles’ rockist ideal of sole authorship, <a href="https://kendisverden.com/mathias-gidsel-alder/">hvor gammel er mathias gidsel</a> modern pop values collaboration, specialization, and production teams adapted to streaming speed, genre fluidity, and global markets.
The data is the data, but you told the story. Thank you for sharing and researching these insights. It was very enlightening. I think in general now, there really arent genres any more in the traditional sense. Due to streaming, things have gone back to the way radio was in the 60s/70s where for the most part everything was played if it was good. People are their own radio stations. Genres are now just a way to describe the sound and train the algorithms, but they dont determine the plays on streaming. They influence it.