Skip to main content

Lennon or McCartney? Scientists Use Artificial Intelligence to Figure Out Who Wrote Iconic Beatles Songs

Do you agonize over the fact that you don’t know for certain who wrote what percentage of your favorite Beatles songs? Do you need to know if a line or phrase is Lennon or McCartney’s before you can enjoy “A Hard Day’s Night,” “In My Life,” and other timeless tunes? Have you lost sleep over the disputed authorship of “Do You Want to Know a Secret”?

I hope not. As Lennon/McCartney themselves wrote, in the end, the songs we love are equal to the love we give the songs…. or something like that. How much we can say with certainty who penned which lyric or melody or played which riff or rhythm part doesn’t add to our emotional experience. But that knowledge does add more to our appreciation than fodder for forum wars or lawsuits.

Pulling these iconic songs into their constituent parts helps confirm our understanding of how those parts contributed differently to making the whole evolve; how Lennon’s directness and simplicity complemented and contrasted with McCartney’s use of “more non-standard musical motifs” and a higher degree of complexity. Or, at least, that’s what an AI found when it analyzed hundreds of Beatles hits in an effort to “build a ‘musical fingerprint’ for each songwriter,” reports Alex Matthews-King at the Independent.

After putting the machine learning algorithm through an initial training phase of “listening” to a complete works, researchers at Harvard “asked” the program to assess “iconic songs, or musical fragments, recorded between 1962 and 1966, where debate rages over who was the major influence.” Much of that debate has been fueled by the songwriters themselves, whose memories in interviews conflict, but who are generally thought to have written most songs individually under their joint songwriting partnership.

The scientists from Harvard and Dalhousie University in Canada were able to gauge with somewhere around 76 percent accuracy whether songs or parts of songs were written by Lennon or McCartney. (Spoiler alert: The AI "was able to identify some, including 'Ask Me Why', 'Do You Want to Know a Secret' and the bridge to 'A Hard Day's Night', as belonging to John Lennon with up to 90 per cent certainty," writes The Daily Mail.) Senior lecturer in statistics at Harvard and paper author Mark Glickman explains the larger purpose of the project to the Financial Times: “Our work is essentially a blueprint for those wanting to follow changes in music over time. Using our machine learning model, you could potentially home in on all the different influences of a given musician.”

If you’re using their work to win arguments, be prepared to explain how the study obtained its results and why they are any more reliable than decades of detective work and expert listening by humans. As a non-statistics person, I’ll leave that explanation to more qualified individuals. I’m satisfied: whether McCartney wrote all of the music for “In My Life” or just the bridge, as Lennon claimed, won’t change the way it moves me one bit.

Related Content:

A Virtual Tour of Every Place Referenced in The Beatles’ Lyrics: In 12 Minutes, Travel 25,000 Miles Across England, France, Russia, India & the US

Watch The Beatles Perform Their Famous Rooftop Concert: It Happened 50 Years Ago Today (January 30, 1969)

A Brief History of Sampling: From the Beatles to the Beastie Boys

Josh Jones is a writer and musician based in Durham, NC. Follow him at @jdmagness

Lennon or McCartney? Scientists Use Artificial Intelligence to Figure Out Who Wrote Iconic Beatles Songs is a post from: Open Culture. Follow us on Facebook, Twitter, and Google Plus, or get our Daily Email. And don't miss our big collections of Free Online Courses, Free Online Movies, Free eBooks, Free Audio Books, Free Foreign Language Lessons, and MOOCs.



from Open Culture https://ift.tt/2LLJS4W
via Ilumina

Comments

Post a Comment

Popular posts from this blog

Ishkur’s Guide to Electronic Music: An Interactive, Encyclopedic Data Visualization of 120 Years of Electronic Music

In a very short span of time, the descriptor “electronic music” has come to sound as overly broad as “classical.” But where what we (often incorrectly) call classical developed over hundreds of years, electronic music proliferated into hundreds of fractal forms in only decades. A far steeper quality curve may have to do with the ease of its creation, but it’s also a factor of this accelerated evolution. Music made by machines has transformed since its early 20th-century beginnings from obscure avant-garde experiments to massively popular genres of global dance and pop. This proliferation, notes Ishkur—designer of Ishkur’s Guide to Electronic Music —hasn't always been to the good. Take what he calls “trendwhoring,” a phenomenon that spawns dozens of new works and subgenera in short order, though it’s arguable whether many of them should exist. Ishkur, describes this process below in an excerpt from his erudite, sardonic “Frequently Unasked Questions”: If fart noises were sudde...
Post a Comment
Read more

A 10 Billion Pixel Scan of Vermeer’s Masterpiece Girl with a Pearl Earring: Explore It Online

We admire Johannes Vermeer’s  Girl with a Pearl Earring   for many reasons , not least that it looks exactly like a girl with a pearl earring. Or at least it does from a distance, as the master of light himself no doubt stepped back to confirm countless times during the painting process, at any moment of which he would have been more concerned with the brushstrokes constituting only a small part of the image. But even Vermeer himself could have perceived only so much detail of the painting that would become his masterpiece. https://www.youtube.com/watch?v=cKaZYTwmjwU Now, more than 350 years after its completion, we can get a closer view of Girl with a Pearl Earring  than anyone has before through a newly released  10 billion-pixel panorama . At this resolution, writes Petapixel’s Jason Schneider , we can “see the painting down to the level of 4.4-microns per pixel.” Undertaken by Emilien Leonhardt and Vincent Sabatier of 3D microscope maker Hirox Europe ...
Post a Comment
Read more

Howard Zinn’s Recommended Reading List for Activists Who Want to Change the World

Image by via Wikimedia Commons Back in college, I spotted A People’s History of the United States   in the bags and on the bookshelves of many a fellow undergraduate. By that time, Howard Zinn’s alternative telling of the American story had been popular reading material for a couple of decades, just as it presumably remains a couple more decades on. Even now, a dozen years after Zinn’s death, his ideas about how to approach U.S. history through non-standard points of view remain widely influential. Just last month, Radical Reads featured the reading list he originally drew up for the  Socialist Worker , pitched at “activists interested in making their own history.” Zinn’s recommendations naturally include the work of other historians, from Gary Nash’s Red, White and Black: The Peoples of Early America (“a pioneering work of ‘multiculturalism’ dealing with racial interactions in the colonial period”) to Vincent Harding’s There Is a River: The Black Struggle for ...
Post a Comment
Read more