Harmony that's Too Stable? Say It Ain't So!

The other day, I was listening to Weezer’s first album - the self-titled “Blue Album”. It’s one of those records where, each time I hear it, I like it more. I just love what this record achieves sonically and culturally. Each time through, songs like “In the Garage” pack a weightier emotional punch. Each time through, I’m just more glad this record exists in the world.

But I’m still a theory nerd. And this time around, when I got to “Say It Ain’t So”, something unusual caught my attention: a G chord that contains both a B♭ and a B♮.

It happens right at the beginning. First, Rivers Cuomo plays a riff that dances around a C minor chord. Then he plays that odd major-minor G chord:

As if to confirm that he did this on purpose, he then plays a riff that goes D-G-B-B♭, sustaining that B♭ even as the lower notes of the A♭ chord enter:

In the charming video, he smiles and looks proud as he plays this atypical bit, as if he’s saying, “Yeah, I just did that. And I meant it.” It’s as if he’s asking us to notice that this chord carries extra meaning.

Weezer is known for relatively straightforward power pop songs; any complexity tends to lie in the attitude and cultural meaning, not the notes. But Rivers Cuomo was no stranger to more elaborate music. He began as a metal guitarist who loved complex instrumental music. In interviews, he talks about how starting Weezer meant suppressing his experience with more complex music. And in 1995, a few months after “Say It Ain’t So” was released as a single, he went off to study classical music at Harvard.

So I had to ask myself - what is Cuomo doing here?

• Is this chord a G major, a G minor, or something else?
• What does this B♮-B♭ conflict mean?
• Putting on our best analyst hat: might this dissonance kick off a journey, a story that’s traced throughout the song?

Let’s explore!

Minor-major conflicts in rock

Before we explore the major-minor conflict in “Say It Ain’t So,” let’s get some vital context in three areas:

1. blue notes in rock
2. a technique American alt-rock bands used in the 1990s to mix minor chords into songs in major keys
3. a different technique used by 1990s Britpop bands

The blue note

The blues is a major influence on rock. Part of why rock is so different from classical music is that the blues is so different, drawn from sources remote from Western Europe.

The musical language of the blues has a completely different sense of major and minor modes than classical music. Although everyone knows the blues when they hear it, the style defies theorists’ efforts to define it. Some people distinguish major and minor blues scales, tying them to two commonly used pentatonic scales. Although I think this idea is somewhat reflected in actual blues and blues-influenced music, I’m not convinced that such music uses these two scales distinctly, and not everyone else is either. For example, in this article, Ethan Hein proposes the idea of a blues tonality. (See the same article for a nice summary of many other theories.)

People do agree on at least one thing: the blue note is a defining feature of the blues. Briefly, blue notes are pitches that lie ambiguously between two pitches of the European twelve-note scale. The blues third is one famous case. Once again, people don’t quite agree on what this third is, except that its pitch is flexible, roaming in the space between a major third and a major second. Sometimes it lies between the major second and the minor third, but it spends most of its time between the minor and major thirds. On a piano, this is often expressed by playing both of those notes simultaneously, or sliding between them. In the key of G, those notes would be B♭ and B♮, and that might sound something like this:

Perhaps because the blues third combines aspects of the major and minor third, blues-influenced music that’s in a conventional major key may well use both types of third - interchangeably, at different times for different purposes, or mashed together simultaneously. This would never have happened in traditional classical music, where major and minor keys were expressed distinctly. But ever since jazz found its way into popular music in the early 20th century, the blues third has colored the major scale with a characteristic bluesy sound. Jimi Hendrix combined major and minor thirds neatly by playing 7(#9) chords in “Purple Haze”:

This sound helped define an era, and now rock fans call it the Hendrix chord. But this combination of minor and major third shows up all over the vast landscape of music influenced by jazz and the blues. It’s even in the organ accompaniment of Henry Mancini’s jazz-influenced “Baby Elephant Walk”:

When the clarinet comes in with the tune, notice that tune contains an A♮ with a grace note A♭. That’s another common way to emulate a blues note.

Sometimes rock songs create hooks by juxtaposing elements from different influences, using contrast and surprise to make music catchy. And one convenient technique involves contrasting blues-influenced harmony with harmony that stems from other influences - classical music, the harmony used in pre-rock 20th-century pop, and so on. Often the blues influence manifests as minor-key tonality, and the non-blues influence results in major-key tonality. Put those together, and you get a song that contains sharp, attention-grabbing major/minor contrasts yet retains coherence.

Major/minor mixture in alternative rock

Classical music theory includes the notion of mixture - when a composition in a major key borrows a chord from the minor mode, or vice versa. Although the effect in rock isn’t quite the same, the concept still resonates in our ears.

In this article on “Old Town Road”, we mentioned that mixing a bit of minor into a major key was a characteristic sound of early-1990s alt-rock. Nirvana’s album Nevermind is drenched in it. Listen to a song like “Lithium”. Often, that moment where you feel that something catchy has happened is a case where the bluesy minor mode has peeked its head into a song in major.

“On a Plain” is similar. Its verse starts like this:

The song’s clearly in D major. But the third chord is not F# minor, the iii chord of D major. Instead, it’s F major, the III chord of D minor. Kurt Cobain’s borrowed a chord from the parallel minor mode. That’s mixture!

To my ear, in this rock context, the mixture is inspired by the blues. The F♮ is an extension of the idea of a blues third. Just imagine how dull this song would sound if the F major chord was replaced with an F# minor:

The song would lose that bluesy, rock edge. It would sound unrealistically happy! Mixture lets Nirvana add bluesy minor thirds to songs in major keys, spicing up their fun rock with soul, preventing them from sounding annoyingly cheerful.

Plus, each time you hear that F chord, it’s a little surprising. It’s a hook. And each time, just when they’ve got you expecting F♮’s, in the following measure, the melody descends to F#. It’s surprising and catchy again. It’s a double hook!

No doubt this album’s popularity and influence propagated this minor-major mixture technique into the rest of the decade.

The Beatles cadence in Britpop

Another type of minor-major combination influences the harmonic language of “Say It Ain’t So”. We’ll call this the “Beatles cadence”.

Why? The Beatles loved to end phrases with a IV - iv - I progression. This article lists examples from “In My Life”, “When I’m Sixty-Four”, and more. It’s another way of inserting a bit of minor into a song in a major key. It may have originally been inspired by the Picardy third of classical music, but in pop it’s taken on a life of its own.

In the 1990s, while alt-rock was popular in America, the U.K. had Britpop. While alt-rock arguably channeled loud rock of the 1970s, Britpop bands liked to evoke the melodic pop of the Beatles. I imagine that’s part of why the IV - iv - I progression became a signifier in Britpop.

The Beatles cadence makes a blatant appearance in Radiohead’s 1992 song “Creep”, a song the members of Weezer knew well. Each time its four-chord progression repeats, you hear a C - Cm - G, which is IV - iv - I.

It’s also prominent in Oasis’ song “Don’t Look Back in Anger”. Listen to the prechorus, at 0:32, on the words “So I start a revolution from my bed”:

It’s deployed again, much more shamelessly, in half-time, at 4:20, at the end of the song:

With this background in mind, let’s dive into “Say It Ain’t So”!

Say It Ain’t So

Like much alternative rock from the 1990s, and like most commercial pop music since then, the verses and choruses of “Say it Ain’t So” share a single chord progression. As the progression’s second chord exhibits both major and minor qualities, we’ll notate it as G(m).

Here’s how that progression sounds with a G major chord, and then with a G minor chord:

In the choruses, the chords are all played as power chords, which means they contain just the root and the fifth, with a missing third.

What key are we in? (Are we in a key?)

Before we proceed, we should decide: what key is this song in? This question isn’t as simple as it sounds.

If you’ve been reading this blog, or, say, Ethan Hein’s much more impressive one, you’ve heard that a pop song doesn’t need to be definitively in a single key. This is a departure from classical music. Classical music from the 17^th to 19^th centuries was fundamentally based on the V⁷-I cadence. Whole compositions explored the journey from tonic to dominant and back to tonic. Even pieces that shifted keys frequently (like the operas of Wagner) were generally at any point either in a certain key or in transition.

In contrast, pop songs simply often lack the mechanisms that classical music used to strongly define keys. Especially these days, pop tends toward repeating chord progressions, which may not be directional at all. Often they just express a cycle.

That said, tonality is often strongly defined in heavier rock music, the kind you want to bang your head to. These songs revel in hitting chords hard, and this added weight can help define a key. For example, the verse of All Time Low’s “Monsters” begins with three heavy Cmi power chords, then quickly settles into a C minor tonality anchored by slow bass hits. Similarly, Queens of the Stone Age’s epic “No One Knows” opens with four C minor chords before launching into a verse whose C minor is defined by repeated low bass C’s (again) paired with a heavy C minor guitar riff. Later, at 2:58 in the track, the whole band pounds the C tonality into our head by playing repeated C power chords.

Defining the key too strongly and precisely does carry risk. Music needs tension to maintain energy. Without mystery, music may become sluggish, tiresome; it can get old quickly. Ideally a song that sits too squarely in a particular key will contain an extra element that introduces dissonance and surprise, giving us something to listen for, keeping our attention. And that’s just what happens in “Say It Ain’t So”.

Weezer’s music isn’t as heavy as that of All Time Low or Queens of the Stone Age, but sludgy rock is certainly an influence on their first records. And, as in the songs we just mentioned, “Say It Ain’t So” does define its key strongly. This is even true in the quiet introduction, where the first vocals, on the words “Oh yeah”, go G-F-E♭:

This moment is our first hint that we’re in the key of E♭. In E♭, G-F-E♭ are scale degrees 3, 2, and 1. Such a 3-2-1 melodic pattern, descending to the first scale degree, would help define a key strongly. Here’s how G-F-E♭ would sound over an E♭ chord:

By the time the descent settles on the first scale degree, you feel like you’re at rest, like you’ve established that E♭ chord.

Of course, in this case, the G-F-E♭ happens over the C minor chord that begins the Cm-G(m)-A♭-E♭ chord progression, not over an E♭ chord. But then there are no further melodic notes until the progression completes, at which point the melodic G-F-E♭ reoccurs on the words “all right”. The E♭ persists in our ear through the gap, hinting that the tune prolongs E♭ throughout, like this:

The first verse makes the case for E♭ more strongly. Two of its two-bar melodic phrases end in E♭ over an E♭ chord, and the other phrase ends on a strong E♭ that’s ornamented by a drop to B♭, the fifth of the E♭ chord. While I play the first verse, listen for those descents to E♭.

The melody varies in later verses, but its phrases consistently descend to an E♭ pitch over an E♭ chord. These powerful cadences to the first scale degree strengthen the E♭ tonality. The only element in the melody that contradicts these restful cadences is the opening B♭-C-B♮, which creates a strong conflict between B♭ and B♮ - and which we will discuss below.

The chorus drives home the case for E♭. The B♭-B♮ conflict of the verse is absent. And the melody arrives at the E♭ by stepwise motion from two directions. The first phrase gets there by descending G-F-E♭, another 3-2-1, with a few extra notes ornamenting the F:

A few seconds later, the chorus' last phrase comes at the E♭ from below, a pop twist on a classic 7-8 cadence, the 7 (a D) appearing over the IV chord, not the V chord we’d find in classical music.

At this point, the instruments pause, making this moment feel like a true cadence.

In the second chorus, the guitars hit a very loud E♭ after the first three Cm and Gm chords. This has the useful effect of defining the E♭ tonality even further while simultaneously introducing some much-needed dissonance.

All these cadences and harmonic stability could make the song boring! Fortunately, another force adds conflict and energy: the conflict between B♭ and B♮. Now we’re ready to investigate that mystery by tracing the journey of the mysterious G(m) chord.

The mysterious G(m) chord

Now that we’ve explored its context, let’s return to the question of the modally ambiguous G(m) chord. We’ll trace its journey through the song’s opening, verse, and chorus.

The opening

The first few seconds of the song showcase the striking dissonance and ambiguity of the G(m). The second chord in the entire song, it contains both a B♭ and a B♮. Next, Cuomo shows that this was no accident, no wrong note, by playing a dissonant D-G-B-B♭ riff:

Right away, we know this song is going to be different. It will be a Weezer song from the blue album period, a song that’s built from conventional rock materials, but that will always stay a bit off-kilter. Without the dissonance, we’d simply have a standard rock song:

It’s hard to play this chord at all on a guitar in standard tuning. Ryan Lendt helpfully points out that Cuomo tuned down a half-step, leaving the dissonant B♭ conveniently on an open string. All he had to do was play an A♭ major chord while keep his finger off the top two strings, leaving the B♭ string open.

Why did Cuomo play it? Probably it just struck him as something odd and attention-getting, something that expressed an emotional state. But it’s essential to the song’s success. In the previous section, we talked about a danger inherent in songs that define their key strongly, with cadences that resolve to the first scale degree: they can lack tension, sapping the energy of the music. This danger is present throughout the Blue Album, where strong cadences are much more common than they are in other rock. Listen, for example, to the ends of the choruses of “In the Garage” and “Buddy Holly”. Both feature full stops after a descent to the first scale degree on a I chord. To keep its energy and flow, “Say It Ain’t So” needs help! This comes from the ambiguity of the G(m) chord’s simultaneous minor and major quality, and the concomitant conflict between B♭ and B.

At 0:13, the beat comes in, along with a rhythm guitar playing Cm - G - A♭ - E♭ chords on offbeats. Yes, the rhythm guitar plays a conventional, good old-fashioned G major chord. This begins to make the case that the G(m) chord is actually a major. Fortunately, each time the rhythm guitar plays a G major, the lead guitar deploys that D-G-B-B♭ riff and throws us back into ambiguity.

At the same time, we hear a small vocal melody that goes G-F-E♭ on the words “oh yeah”. As we described above, this 3-2-1 descent to the first scale degree reduces tension. Fortunately, each time it happens, the lead guitar immediately counteracts the undesired tranquility by playing the D-G-B-B♭ riff.

The introduction ends with a full stop on the E♭ chord that closes the progression. This is preceded as always by the A♭ chord, creating what feels like a plagal IV-I cadence. (More on these later.) Something needs to restart the energy!

The verse - mixture

Fortunately, the verse wastes no time intensifying the battle between B♭ and B♮. The first melodic phrase consists of the notes B♭ - C - B♮:

This tune is quite unusual. It’s highly chromatic! It describes an abrupt move from the minor mode implied by the Cm chord and B♭ pitch to the major mode of the G chord and B♮.

After this striking beginning, the verse becomes conventional, descending another 3-2-1 scale degree progression over an E♭ chord. Thereafter it stays safely away from this B♭-B♮conflict with a couple of phrases that descend to E♭ or B♭.

That B♭-C-B♮ is a small element of dissonance stuck into an otherwise consonant bit of music, just as the G(m) chord and D-G-B-B♭ were in the introduction. Once again, it’s just enough to keep the tension going, by forcing the B♭-B♮ conflict back into our faces.

So, why does the melody contain both a B♭ and a B♮? The B♭ is there because the song is in E♭, and B♭ is the fifth scale degree of E♭. Almost any song in E♭ will be drenched in B♭’s. But - why B♮?

B♮ enters the picture as the third scale degree of the G major chord over which it occurs. But, then - why is there a G major chord in the key of E♭? After all, in this era of pop music, most songs used a simpler harmonic palette, generally focusing on chords drawn from a single scale. This would lead us to expect a G minor chord in “Say It Ain’t So”, as part of a Cm-Gm-A♭-E♭ chord progression, vi-iii-IV-I in the key of E♭.

Remember our discussion above about mixture and the Beatles cadence? I think that this G major chord reflects both. But because this is pop music, it’s more loosely constructed than it would be in a classical composition. Let’s look at this in detail.

Let’s start with mixture. To me, this resembles a very mild case of mixture, something that was also common in classical music: the applied dominant. In classical harmony, you couldn’t just shove a III chord into the music, because it contains a pitch that’s not part of the key. But you could use this III chord if it was directly followed by the vi chord. The III would function as a local dominant of the vi, forming a miniature V-i in a tiny vi bubble. To make this clear, you’d notate the III as V/vi.

Here’s an example. Let’s say (conveniently) you were in the key of E♭ major. You want to use a G major chord, and you follow it directly with a C minor chord. The G major resolves to C minor in a V-i pattern. G major - C minor is then V/vi - vi. Here’s how that sounds:

Unfortunately, the G major in our song is followed not by Cm, but by A♭. Our applied dominant narrative requires another subplot borrowed from classical music: the deceptive cadence. This is a cadence in which the first chord leads to a second chord that’s unexpected. For example, instead of V⁷ - I, it might go V⁷ - vi. Here’s a standard V⁷-I cadence and a deceptive V⁷-vi cadence side by side:

This deceptive cadence doesn’t fully resolve the tension created by the first chord - it’s more a clever, subtle way of extending it.

A deceptive cadence can be combined with an applied dominant. This gives us an intriguing way of hearing the chord progression in “Say It Ain’t So”. The Cm - G - A♭ - E♭ progression can be analyzed as vi - V/vi - IV - I, with V/vi - IV as our deceptive cadence.

In other words, after a C minor chord and a G chord, our ears would expect a return to C minor:

But instead we get the A♭ chord, which then drives us back down to E♭ via a plagal cadence:

Plagal cadences… hmm. What else from this article does that remind us of?

The verse - the Beatles cadence

Aha!

Do you remember Radiohead’s song “Creep”? If you wanted to analyze it using classical music techniques (which you might not want to do), you might find the same combination of applied dominant and deceptive cadence. You’ll recall that its chord progression is G - B - C - Cm. You might interpret the B and C chords as a V/vi followed by a IV. You’d then hear the four chords as I - V/vi - IV - iv.

But then there’s the Beatles cadence. Remember that as each chord cycle of “Creep” progresses to the next, we hear a IV - iv - I progression:

“Say It Ain’t So” does not include a IV - iv - I progression. But it does echo the Beatles cadence. Its first two chords are the same as the final two chords in the example above. I hear the song strongly as twin plagal cadences, one in the key of G, the other in the key of E♭:

The plagal-cadence-like structures carry their traditional emotional weight, lending the song a feeling of resignation, of calm resolution. It’s like we resolve down to G, then creep up a half-step to A♭ before sinking down to E♭, the tonic.

We’ve gone down a bit of a dangerous route here, analyzing rock music like it was classical music - which it is not. This technique can lead to all sorts of fallacies, so it should be used in moderation. Pop music is not classical music, and its harmony works in different ways. But its chord progressions do derive some of their meaning from the same forces that animate classical music. In the right context, a IV-I chord move like A♭-E♭ can carry the emotional and music-energy weight of a plagal cadence.

When you realize that pop treats these classical ideas more loosely, you can hear more of what’s going on. In “Say It Ain’t So”, it’s actually as if the progression is happening backwards. Reversing it, you get E♭ - A♭ - G - Cm. The G now properly fulfills its classical function as an applied dominant to Cm, making the progression I - IV - V/vi - vi:

Let’s crawl back out of the classical harmony rabbit hole and look at this in a more obvious way. Put simply, the G major chord and the resulting B♮ create necessary tension by injecting a note outside of the scale. Whether we call the G major an applied dominant, a borrowed chord, or something else, it’s there to introduce color and variety. It smoothly brings in a fresh chord that contains a new pitch, the ♭6. It gently pulls the music away from the tonic chord toward a different tonal center. And it certainly lends the song energy. Imagine how much less interesting this song would be if it used G minor instead of G major. We’d also need to replace the B♮ in the melody with a B-flat:

Compare this to the actual song:

You can feel how the G major chord makes the tension rise. And how strongly it drives us forward to the A♭ chord, as the B♮ in the G major chord pulls upward to the C in the A♭ chord. And then how somehow it all resolves in the E♭ chord.

Notice too how different this is from “Creep”. The melody of “Creep” contains both the B♭and the B♮, but they’re artfully separated, a smooth consequence of the Beatles cadence:

In contrast, in “Say It Ain’t So”, Cuomo seizes the opportunity to create a shocking, angular, desperate, chromatic B♭-C-B.

Interestingly, the first verse is followed not by a chorus, but by a return to the introduction. Then comes a second verse, which begins with an even louder and bolder B♭-C-B♮ (“flip on the telly”). Only then does the song reach its first chorus.

The chorus

The song falls into the Nirvana/Pixies school in which the verse and chorus use the same chords, but the chorus is much, much louder. The verse doesn’t need to build up to the chorus; instead, the chorus enters with sudden contrast, stomping on the stomp pedal and piling on the distortion.

This chorus lives fully in the land of heavy rock stability. Here, the B♭-B♮conflict is simply absent. The chorus does nothing to resolve the conflict. Instead, it avoids it, acting as if it never existed. The melody and chords contain an occasional B♭, but the B♮ is just missing.

In this land of loud rock certainty, what’s become of our ambiguous G(m) chord? In the chorus, the chords become power chords, minus the third that would pin down their qualities.

During the chorus’ first chord cycle, the melody doesn’t start until the A♭ chord, leaving the power chord flavors of Cm and G(m) to float in an ambiguous space between major and minor:

This melody is an ornamented version of the opening’s G-F-E♭. It’s another cadential figure in the key of E♭. Only the strong rhythmic emphasis on F, the second scale degree, saves it from finality.

As the melody continues through the second progression, it passes through a B♭, a minor third which finally defines the G(m) chord as a minor.

As mentioned above, the melody concludes with something like a full cadence in E♭. And that’s our story. The G(m) chord became a G major in the verse, though coupled with a strong B♭-B♮conflict. In the chorus, there’s no B in the melody, just B♭, shifting the underlying chord to G minor. The tale is complete!

The bridge

By the time we get to the bridge, the story of B♭ and B♮ is over. But Cuomo wants to celebrate the victory of B♭, to revel in the moment. He does this in a big way, by changing the song’s key to B♭ for the entire section. It’s not so unusual for the bridge to move to the V chord, but normally in such cases one feels that the song’s there temporarily, poised to return to a final chorus and the I chord. Here, we land strongly in B♭, with a new chord progression rooted deeply in that key:

In the bridge, when we hear the E♭ chord, it’s now clearly the IV of B♭. And just as the verse and chorus progression includes the borrowed G major chord, the bridge progression incorporates a borrowed chord of its own - a G♭ chord, on loan from B♭ minor. G♭ is a major third down from B♭. This is a mirror image of the G, which was a major third up from E♭. And, just as the G major chord brought the fresh melodic pitch of B♮ into the key of E♭, the G♭ chord brings the fresh melodic pitch of D♭ into the key of B♭:

Of course, the G♭ in the accompaniment is a new pitch as well. At the end of the bridge, this pitch wonderfully finds its way into the melody. And then it resolves chromatically upward to a G♮ as the harmony shifts to Cm, marking the return to the song’s main Cm-Gm-A♭-E♭ chord progression. It’s a retransition worthy of the finest 19th-century classical music:

This is followed by another 3-2-1 melodic cadence in the key of E♭. This one is triumphant, thrilling, so suffused with energy from the recent transition that it needs to be followed by a guitar solo.

From this point forward in the song, the B♭ is ascendant, and the G chord remains consistently minor.

The end

We can now summarize the story of the ambiguous G(m) chord, as well as the subplot which may have been the main plot all along: the battle between B♭ and B♮.

You may think it’s a little much to talk so much about a conflict between two pitches. But in fact, this sort of drama occurs throughout tonal music. Beethoven, in the Scherzo of his Hammerklavier Sonata, plays with a conflict between B♭ and B♮, the very same pitches that Rivers Cuomo played with. At the end of the first movement, Beethoven concludes a phrase with a strong cadence on B♭, capped by two loud B♭’s in octaves. Suddenly, the pianist then plays two very soft B♮’s. Then, two more loud B♭’s - and two more soft B♮’s. The pianist starts playing the main theme of the piece in B♮, slowly, quietly, clearly in the wrong key. This culminates in a crescendo of 21 quarter note B’s - that end dramatically in two definitive B♭’s. The conflict is resolved, and the movement ends as if nothing had happened. Listen to that here.

Back in the world of rock and of Weezer - other songs on the Blue Album play with unusual half-step dissonances as well. Just listen to the lead guitar part in the introduction of “Undone”. Over three of the four chords, it plays D#-A-A#. Over the other chord, it plays E#-B-A#. B and A# - that’s quite similar to the B-B-flat dissonance we’ve discussed here. The highly ambiguous chords the lead guitar forms here deserve an article of their very own, though if you assume that the underlying chords match the F# - B - C# - B progression in the rest of the song, these are surface dissonances - appoggiaturas or incomplete neighbor notes. In other words, the lead guitar doesn’t contradict the fundamental nature of the harmonies.

A two-note conflict in a Weezer song has led us into a deep exploration of chromatic techniques in alternative rock. Hope you enjoyed it! I promise to write something more detailed about how mixture helped make Nirvana’s songs catchy. Soon. Soon. I promise!