First, let’s get this out of the way: the title of this post is a joke. No chord progression has any inherent emotional quality. Musical sadness is a matter of cultural convention, and even within a culture or subculture, sadness is the result of harmony interacting with melody, rhythm, tempo, timbre, phrasing, articulation and other intangibles. The listener produces as much of the music’s meaning as the music does, if not more. That said, everything else being equal, there are some chord and voice leading combinations that reliably evoke sadness in Anglo-American listeners. The saddest chord progression that I know of comes from a short passage near the end of Vasily Kalinnikov‘s Symphony No. 1, 2nd movement. Listen at 6:16.
I mean, right? So the question is, what makes this so sad? Some of it is the orchestration and dynamics and so on. But even if you strum these chords on a guitar with minimal expressiveness, they are still sad. Let’s find out why.
The passage is in E-flat major, but I transposed it into C in the transcription below. All the other transcriptions in this post are transposed into C as well for ease of comparison.
The progression begins with Am, the relative minor chord in C. The next chord is D, which is outside the key of C. It’s a secondary dominant, the V chord in the key of G. In European classical music, this chord would typically lead to G, which would then resolve to C. However, this is not what Kalinnikov does. Instead, he follows D with Fm. This is another chord from outside the key; it’s borrowed from parallel C minor. Kalinnikov set you up to expect a move toward the sharp side of the circle of fifths, but instead he goes several steps over toward the flat side. The last chord is the tonic C. Kalinnikov gives the chords some extra flavor by repeating the notes E and C over all four of them. These notes are chord tones in Am and C, but they create colorful extensions on D and Fm, turning them into D9 and Fm(maj7) respectively.
Let’s look a little more closely at the voice leading in these chords. First, consider the notes I colored blue in my chart. The F-sharp in the D chord moves down a half step to the F-natural in the Fm chord, which in turn moves down a half step to the E-natural in the C chord. Now look at the notes I colored green. The A in the D chord moves down a half step to the A-flat in the Fm chord, which in turn moves down a half step to the G in the C chord. The takeaway here is that if you want to make me feel sad, apparently you should move a minor third down in half steps, and do it at a slow tempo.
Part of the reason Kalinnikov’s chord progression works so effectively on me is the nostalgia factor. After I first heard it, I realized that I was already familiar with the progression, from a few different contexts. I had probably first heard it in “In My Life” by the Beatles, in the second half of the bridge.
As a teenager I thought that this song’s sadness was due to its lyrics, but as an adult I recognize that the lyrics of Beatles tunes are mainly just decoration for the melodies. Here’s my chart.
Kalinnikov’s chords were also familiar to me from Willie Nelson’s recording of “I’d Have To Be Crazy” by Steven Fromholz. I performed this tune many times in my country music days, and I sang it as a lullaby for my kids when they were little.
The chords in “I’d Have To Be Crazy” follow a different harmonic rhythm from the ones in Kalinnikov and the Beatles, and they start on (transposed) C rather than (transposed) Am. Nevertheless, the effect is the same.
The descending chromatic minor thirds in the Kalinnikov chords are also similar to this classic blues trope.
This riff doesn’t have an official name; I call it the descending diminished blues cliche (though I would love to find a catchier name). One of my favorite jazz tunes, “Way Way Back” by Johnny Hodges and Mercer Ellington, uses a variant on the riff, also with Kalinnikov-esque voice leading. Here’s Abdullah Ibrahim’s recording.
Paul Gonsalves recorded a lovely version too, but I can’t find it on YouTube. Anyway, here’s a chart.
While the blues trope is structurally similar to Kalinnikov’s progression, its feel is very different. The blues certainly can express sadness, but that isn’t the main feeling I get from it. I think of the blues as being more about overcoming or coping with sadness than evoking it in the listener.
The Kalinnikov progression is also related to the “Beatles cadence,” more accurately known as a combination major/minor plagal cadence. You can hear it at around 1:00 in “If I Fell,” in the bridge.
Check out the line “but I couldn’t stand the pain.” The word “pain” lands on the F chord, the IV chord in C major. In the next line, “and I would be sad,” the word “I” lands on Fm, the iv chord from parallel C minor. The active musical ingredient here is the descending voice leading from the A in the F chord to the A-flat in the Fm chord to the G in the C chord.
There’s another close relative of the Kalinnikov progression in “Can’t Take My Eyes Off Of You“, most famously recorded by Frankie Valli.
People my age are likelier to prefer the Lauryn Hill version:
Either way, the opening chords are the same:
Here we have the same D chord as in the Kalinnikov progression, but it goes to F, not Fm. There’s the chromatic descent from the F-sharp in the D chord to the F-natural in the F chord to the G in the C chord. However, rather than having the A in the D chord fall to A-flat, it carries through to the A in the F chord. The end result is still wistful, but definitely not sad.
Aside from the descending chromatic voice leading, the thing that makes the Kalinnikov progression so magical is the way that it sets up and then subverts your expectations. In European classical tradition, a D chord in the key of C makes you expect to hear a G chord, because D is the dominant chord in the key of G. Then after D resolves to G, you expect G to resolve back to the original tonic, C. It sounds fine, but it’s not very exciting.
A jazz musician might think of these chords in terms of scales. In C major, the scale implied by D7 is D Mixolydian, which contains the same pitches as C Lydian, the brightest of all the diatonic modes. Landing on the G7 chord puts us back in C major. There’s a bit of tension from the F rising up to F-sharp and then falling back to F, but basically, all is sweetness and light.
This is not what happens in the Kalinnikov progression. Instead of trotting obediently around the circle of fifths like you expect, the D7 unexpectedly resolves to Fm instead. This surprising chord implies F Dorian, the same pitches as C natural minor. So you go from the bright C Lydian sound (C, D, E, F-sharp, G, A, B) to the much darker C natural minor sound (C, D, E-flat, F, G, A-flat, B-flat.) Instead of moving from the brightest scale to the second brightest, you have shifted abruptly into a much darker harmonic environment. Sadness is that much sadder if you were expecting happiness.
I did a remix of the Kalinnikov chords with some samples of Charles Mingus and the Detroit Emeralds, enjoy.
See also the happiest chord progression ever.
I first became familiar with a very similar progression in “When I Was Your Man” by Bruno Mars. …all you wanted to do was (Am) dance, (D) now my baby’s (F) dancing, but she’s (Fm) dancing with another (C) man.
I also am not sure that it’s entirely culturally dependent whether or not a song is sad…here’s an Armenian song that has always sounded sad to me, despite the fact that it doesn’t really follow the same conventions as Western Classical or Pop/Rock music.
This sounds quite similar to Western classical to me, aside from some Middle Eastern sounding flourishes here and there. I don’t know what meaning the song would have to an Armenian person, but it certainly has many of the signifiers of “sadness” by Western European standards (slow tempo, minor key, quiet dynamics etc).
Seems a lovely counterpoint (pun intended) to Alex Ross’ piece on the chacona/lamento. https://www.therestisnoise.com/chacona/ Throwing it into the class I teach that draws on that next time I do it :)
“No chord progression has any inherent emotional quality. Musical sadness is a matter of cultural convention, and even within a culture or subculture, sadness is the result of harmony interacting with melody, rhythm, tempo, timbre, phrasing, articulation and other intangibles.”
Hmmm… have you ever considered that before music is a function of “culture” that it is a sound? Do you think that sounds can not possibly have an emotional effect, directly from perception? Is it possible that in the evolutionary development of sound perception, there might be emotional tools for survival reactions? Might one watch out that the analytical hammer turns the world into a nail?
I have considered this. Sure, sounds can have an emotional effect, but the nature of that effect will heavily depend on culture. It’s impossible for an adult human to have an experience that’s unmediated by culture. The idea that there is some kind of universal evolved music cognition does not withstand scrutiny. The same person can experience the same piece of music as meaning very different things depending on their life experience to that moment. Two different almost identical people can experience the same piece of music very differently. I find Aphex Twin’s Selected Ambient Works Volume II to be relaxing and peaceful; my wife finds it terrifying. There’s a klezmer tune called “Der Gassen Nigun” that sounds to me like a funeral dirge, but it’s actually a wedding march, and that’s how my own Jewish great-grandparents would have experienced it. There are no universals among perceptions of musical meaning. Humans don’t even all consider octaves to be equivalent, much less all agree that minor chords are sad or whatever.
Does a few personal anecdotes constitute a proof? Do not exceptions, sometimes, “prove rules”? Is Homo sapiens never guilty of “perverse” behavior (to explain exceptions)?