Support independent, student-run journalism.

Your support helps give staff members from all backgrounds the opportunity to conduct meaningful reporting on important issues at Stanford. All contributions are tax-deductible.

Songtology: What’s in a song?


For how important music is to us, we don’t understand much about it.

Some people engage with music actively, choosing to do nothing else while they listen to it, and others use it passively as a soundtrack to accompany the events of their lives. Some people turn to sad songs when they are sad and others avoid them entirely. We feel a profound sense of ownership over the music we discover and show to our friends.

What accounts for these feelings? And how are they changing with the digitalization of music, the rise of streaming services, and global accessibility? I hope to explore some of these questions in a series of columns centered on the philosophy of music: Songtology. This week, I’ll begin with a metaphysical, foundational question: What allows us to identify one particular song as that song?

If we strip a song into its constitutive elements, we find that no single part is essential to its identity. Let’s start in an obvious place: the lyrics. If I strum the chords of “Blackbird” while humming the melody and ask you what song it is, you wouldn’t say, “It’s something similar to ‘Blackbird’ but missing the lyrics!” You would simply say I was playing “Blackbird.” You might call it an instrumental version, but even then, the essential Blackbirdness of what I’ve played remains. Words alone are not necessary for a song to maintain its identity.

What about the melody? Even if the words aren’t spoken, a melodic instrument that closely follows the pattern of the lyrics implies them. This too is inessential; people often perform stylized versions, or covers, of songs in which they offer a unique interpretation of the melody, adding vocal riffs, lingering on certain notes and delaying central, recognizable rhythmic and tonal aspects that differ wildly, almost unrecognizably, from the original. Even beyond intentional deviation from a melody, a band could perform a song and get key notes of the melody wrong accidentally. You could say they played badly, but you wouldn’t say that they did not play the song. 

What about the range of notes available to deviate from the melody? A specified key signature and chord structure underlie any musical improvisations that occur in the song and determine the range of choices available when creating a melody in the first place; in a sense, they are the backbone of the composition. But nothing crucial is here, either; bands change the keys of songs all the time to accommodate a particular singer’s voice, and we rarely even notice. An increasing number of popular “reharmonizations” of songs frame the same melody over different chords (Jacob Collier’s version of “Close to You” is an interesting example), and jazz artists have been reharmonizing for as long as jazz has been around.

Rhythmic motifs can play a large role in the identity of a song but often change when other artists perform the piece. In live performances, bands rarely play a song at the speed it was initially recorded without accidentally speeding up, and hip-hop artists often slow down live performances to accommodate particularly wordy verses.

By referencing the myriad covers and live performances of a song, we can easily dispatch arguments of necessity for any single dimension that we suspect is necessary for a song to be considered itself. In order to conceptualize this idea of a cluster of non-essential qualities that all add up to one thing, we can consider the philosophical example of Theseus’ Ship, a theory of metaphysical identity in the Western philosophical tradition that dates back to Plato.

Imagine a famous ship — let’s call it the Leland — that is moved into a museum for display. Over time, certain planks that constitute the hull of the ship begin to rot. After one plank is replaced, the Leland is still surely the Leland, no? How about after half of the boat is replaced? Or the whole thing? At some point, one might feel compelled to say the Leland was destroyed and that the new boat is merely a replica of the Leland. Others might say it is in fact still the same boat.

We can think of each element of a song as a plank and the overall boat as The Song. If we replace elements of a song with similar elements — an electric guitar sound for a distorted synth sound, a melody for a similar variation of the melody — it would be like replacing old planks with very similar planks. Changing one of these elements would not likely provide sufficient reason to doubt that the boat is no longer the boat, or that the song is no longer the song.

But if we make enough individually innocuous changes, then we start to grow suspicious. If the melody is different, the chords are different, the lyrics are missing and the rhythmic elements have changed entirely, then we can be more certain that the new song we listen to is different in an important sense from the original, even if it is derivative from it.

The analogy accounts well for larger changes. If I exchange the ship’s old wooden rudder for a massive speed boat motor, I’ve done something to substantially alter the Leland. We might think of it now as “The Leland but with a speedboat motor.” In the same way that reharmonizing a song feels more substantial than playing a slightly different drumbeat, we could say that a particular version of Song X is “Song X but reharmonized.” Its identity is partially maintained.

The Ship of Theseus thought experiment is helpful, but not holistic, and has left us with more questions: what does it mean to be an interpretation of something versus an adaptation, version, derivative work or instance? Even though no single element is essential to the song’s identity, is reference to a certain element, like the melody, essential? It also only helps us answer questions of what constitutes the ship, not where the ship itself resides — in live performances? On a record? Or as an abstract object in our imagination? In the next column, I will discuss this question of where.

Philosophical inquiry never promises answers — only questions. But the kind of questions we ask do significant work in clarifying the relationship we have with abstract concepts. In the quest to define the range of options we have when it comes to our relationship with music, asking questions that illuminate the structures we navigate are answers in their own way.

Contact Tristan Wagner at triwag ‘at’