It’s not often that an article about science and music gets as much attention as “Anatomy of a Tear-Jerker,” a column in the Wall Street Journal that purports to explain why Adele’s hit song “Someone Like You” makes people cry. The story circulated widely on the internet and was even picked up by NPR’s All Things Considered. It’s easy to understand the appeal of an article like this; naturally we’re curious about the origins of our emotions. And it gives people a rare chance to feel objectively “right” about their musical preferences: it’s not taste, it’s science!
Unfortunately, the article is marred by a number of scientific, musical, and aesthetic misconceptions, some glaring and some more subtle. I know I’m not the only musician who is frustrated to see errors like this in a major publication. The scientific and music theory mistakes are fairly easy to quantify. The aesthetic issues are actually larger and more urgent, but harder to pin down. Let’s start with the quantifiable errors.
Appoggiaturas, Tears, and Chills
The article makes much out of appoggiaturas, as if they were some kind of magic bullet for generating tears. But the term appoggiatura has different meanings depending on context, and the article confuses and conflates them. It’s true that the appoggiatura is a term for a specific kind of ornament, with a particular method of notation, that appears most often in Baroque and classical music. But while the article continually refers to appoggiaturas as “ornamental notes,” the study by John Sloboda it references—”Music Structure and Emotional Response: Some Empirical Findings” (Psychology of Music, October 1991, 19:110-120)—is actually discussing another kind of appoggiatura altogether. What Sloboda calls a “melodic appoggiatura” is a more general term for a certain kind of dissonant note, or non-harmonic tone. Non-harmonic tones are omnipresent in Western tonal music. Without getting too far into detail, the melodic appoggiatura differs from most other non-harmonic tones because the dissonance occurs on a strong beat and therefore receives more emphasis than usual. (For more information, see Edward Aldwell, Carl Schachter, and Allen Cadwallader. Harmony & Voice Leading, 4th Ed. Cengage Learning, Boston, MA, 2010.) But the article doesn’t even assert that “Someone Like You” contains appoggiaturas; instead, it “is sprinkled with ornamental notes similar to appoggiaturas.” In other words, it uses non-harmonic tones, just like nearly every other melody in Western music.
In a recent addendum to the NPR piece, Rob Kapilow purports to correct the definition of appoggiatura, and points out a few accented dissonances in “Someone Like You.” However, the main problem is not with the definition, but with the implication. If we generalize the appoggiatura to mean simply “a fleeting dissonance,” I question whether something this commonplace can really be considered a meaningful feature by itself. It’s a little like singling out a letter of the alphabet for similar properties. In fact, Sloboda himself questions the meaning of his results. I’ll dig into this more later, but first let’s look at a few more points from his study:
1) The study’s participants were only asked to specifically identify which passages caused an emotional response, so we have no idea how many passages containing appoggiaturas didn’t cause an emotional reaction.
2) Since the appoggiatura is a fairly common musical device, it’s an almost absolute certainty that there were many similar passages that didn’t elicit an emotional reaction.
3) Even within the study’s small sample size of 38 passages, there were two tear-causing examples that didn’t contain appoggiaturas.
Therefore, while we see a general tendency for appoggiaturas and tears to be associated, we can’t say that appoggiaturas are a necessary or sufficient condition for causing tears.
The article also discusses “chills” caused by sudden changes in register, timbre, and harmony, and equates tears and chills as if they were part of the same emotional response:
Chills often descend on listeners at these moments of resolution. When several appoggiaturas occur next to each other in a melody, it generates a cycle of tension and release. This provokes an even stronger reaction, and that is when the tears start to flow. [emphasis added]
A quick glance at a table of music-structural features from Sloboda’s study reveals a more complicated picture:
This result seems to confirm that chills (or “shivers”) tend to be associated with sudden changes in harmony or texture, though again, we don’t know how many sudden changes didn’t trigger an emotional response. However, tears and chills are categorized as completely different physical reactions, generally triggered by different musical features. Another study referenced in the article (Martin Guhn, Alfons Hamm, and Marcel Zentner, “Physiological and Musico-Acoustic Correlates of the Chill Response,” Music Perception: Vol. 24, No. 5 (June 2007), pp. 473-484) deals only with chills, and doesn’t mention tears or crying at all. So, even if the octave leap in Adele’s voice at the chorus of “Someone Like You” causes listeners to experience chills, there is no evidence this would have anything to do with making people cry.
Emotional Responses to Music are Learned
So if appoggiaturas and surprise shifts don’t account for the tearjerking qualities of “Someone Like You,” what does? Unfortunately (for journalists) or fortunately (for musicians and researchers), the answer is not easily summarized. First of all, it’s important to state the obvious: not everyone has the same reaction to the song! In the wake of the article, discussion on websites like Metafilter revealed not only agreement, but also substantial bemusement and dissension:
“Huh…it doesn’t really stand out to me as a song.” —droplet
“There’s something trite about the melody that irks me.” —timsneezed
“I’d like some scientific explanation of why that song does nothing for me.” —Wolfdog
I bring this up not to disparage the song but to point out that the emotional reaction to it is far from universal. Are these people just cold, emotionless robots? Well, no. They just have emotional reactions to different songs. This suggests something more complex than a hard-wired biological reaction. It suggests a network of cultural and personal factors; in short, it indicates that our emotional responses to music are learned. This is confirmed by the backgrounds of the respondents in Sloboda’s study, who were mostly experienced performers. Sloboda himself acknowledges this quite directly:
It is clear that the ability to experience these responses in connection with specific musical structures is learned. This is because (a) the responses are not shared by young children and people from different musical cultures; (b) the structures mediating these responses are often only perceptible in terms of a musical syntax…and (c) the emotional response to a piece of music can grow during repeated exposure to the same piece, as one discovers more of the subtle structural features….For these reasons, the present results can lead to no strong claims about the extent to which the pattern of response reported here will be replicated in less experienced musicians. [emphasis added]
(One wonders why Sloboda doesn’t address this in the NPR piece; possibly, it didn’t fit into a soundbite.)
What kind of learned musical structures do we possess that would help us understand “Someone Like You”? Well, as a culture, we are well-versed in popular music from the 1960s to the present. “Someone Like You” uses a looped chord progression (I-V-vi-IV) that is very common in popular music, and often associated with other melancholy songs, as another Metafilter user pointed out:
“When I hear the chord progression in the chorus, my brain makes a mental segue to U2’s ‘With or Without You,’ another singalong anthem.”—carter
It’s interesting that the prominence of this chord progression is fairly recent; it’s not hugely important in Western classical music, for example. The Beatles’ “Let It Be” is one early example of a pop song built on this progression. Another thing that lends this progression a bittersweet quality is its ambiguous tonality, constantly shifting between a major key and its relative minor. But it’s important to note that the association of major and minor keys with “happy” and “sad” feelings is culturally learned, too.
Typical vs. Exceptional
There’s one final piece of the puzzle missing. Experiments like Sloboda’s are effective when identifying structural features that are typical, that is, features that are commonly found in a large variety of musical examples. What they are not capable of is locating unusual features: that is, what makes a piece of music unique or special. But great songs, songs that we love, are by definition exceptional—there’s something about them that other songs don’t have. Otherwise every song with the same basic features would evoke the same exact reaction, which is clearly not the case.
If “Someone Like You” were just a rote rehash of an old chord progression, it’s unlikely that it would have the same impact. But there are a number of subtle structural features that stand out on further inspection. One Metafilter user discusses some of these features:
“The song is in 4/4 time and this pre-chorus starts with 4 bars, as most pop songs would normally unfold. But it ends with an extra bar of 2/4 added on to it, where Adele appropriately sings ‘it isn’t over,’ thus yearningly stretching out the phrase. [At the end of the chorus] the missing half of the earlier bar that should have been 4/4 but was changed to 2/4 is ‘paid back,’ at which point we hear the piano linger on for an extra inserted bar of 2/4 just before she starts singing the next verse. Time lost and then remembered.” —colie
In other words, the song deftly plays with our expectations of rhythm and time. But these expectations are malleable; if every songwriter picked up on this trick and made it utterly commonplace, it would no longer be quite as effective. Music is always a moving target.
This brings me to the aspect of the article that I find most offensive, the implication that music is like a science of emotional manipulation through sound, and that it’s as simple as applying a “formula” to achieve commercial and artistic success. Not only is it belittling to musicians and listeners everywhere, it also implies a very narrow view of musical craft. I want to strenuously argue for the value of music that doesn’t necessarily cause tears or chills. The burden on music to communicate certain, specific emotions can be oppressive. Composers like Igor Stravinsky and John Cage, for example, found it so inimical that they rejected the idea of self-expression entirely. Cage in particular describes an artistic situation where a lack of shared cultural context seems to preclude any kind of communication: “No one was understanding anybody else. It was clearly pointless to continue that way, so I determined to stop writing music until I found a better reason than ‘self expression’ for doing it.” (Richard Kostelanetz, ed., Conversing with Cage, New York: Limelight, 1988.)
Of course, not everyone will want to follow Cage’s uncompromising path. But this is all the more reason why we should want an open musical world. I suspect that everyone who makes music has an individual, personal goal for what they hope their music will accomplish, and this is why we create in the first place. The kind of music we want to hear doesn’t always exist yet, and no formula already known to us will bring it to life.
Isaac Schankler is a composer and improviser based in Los Angeles, California. He is the artist-in-residence at the University of Southern California’s Music Computation and Cognition Laboratory, and an artistic director of the concert series People Inside Electronics.