Why Modern Musicians Love The Four-Chord Loop

Why Modern Musicians Love The Four-Chord Loop

this video is sponsored by Skillshare! hey, welcome to 12tone! a couple years back,
I made a video called How To Write A Four-Chord Loop, where I took a look at one of the most
maligned aspects of modern popular music and explained why they’re actually really interesting
and useful. it’s a good video, and I stand by it, but
recently I was reading a book by theorist Philip Tagg, who has a completely different
way of looking at four-chord loops. it’s a really interesting approach, so I figured
that, since thousands of hit songs have used these loops, I can probably justify making
a second video about them. let’s go! (tick, tick, tick, tick, tock) to understand Tagg’s approach, it might be
helpful to look at a slightly simpler concept first: the chord shuttle. this is his term for a progression that just
goes back and forth between two chords. his primary example is Pink Floyd’s Great
Gig In The Sky, which spends over half its runtime alternating between Gmi7 and C7. now,
one of the first questions a theorist might ask when looking at this groove is, which
key is it in? and the answer isn’t easy. Tagg proposes four possible interpretations:
first, it could be in something like G minor, in which case this is I going to IV. that
means we’re starting on the root and then venturing off to a less stable sound. the second option is that we’re in C, making
this a sort of V-I, with our resting place happening on the second chord. third, it’s
possible neither of them is the root: jazz players might recognize this progression as
a classic II-V, implying a resolution to F that we just never actually get. and finally,
we could be in Bb, because that’s the chord the rest of the song mostly resolves to, in
which case this is VI and II. so which of those is correct? well, Tagg argues that none of them are. asking what key we’re in is asking the wrong
question: by looking for a single tonal center, we’re assuming the progression has a specific
structure, one with a clear, linear direction, but if you listen to the song… it kinda
doesn’t. to my ears at least, neither chord really feels like it wants to go anywhere. to quote Tagg, the chord shuttle is “not a
place you pass through en route to another destination: it is itself somewhere to be.” that is, what matters here isn’t what each
chord wants to do next. what matters is what those chords are doing
right now. we’re not in G, C, F, or Bb: we’re in the
harmonic space that exists between the chords Gmi7 and C7. that’s it. the two chords are, effectively, opposite
poles within a single tonality, like a sort of musical superposition between the two roots. this puts Tagg in sharp contrast to the more
classical model of functional harmony. when I talk about this on the channel, I usually
say it’s the idea that different chords in a key have different functions, or jobs to
do. that is, they tend to behave in specific,
predictable ways: tonic function chords provide a sense of rest, subdominant chords disrupt
that rest, and dominant chords point you back to that rest. in classical harmony, we expect these three
functions to fit together into strong patterns of tension and release, with clear resolutions
to mark out the boundaries of our harmonic home. but I’ll take this a step further: in some
kinds of classical theory, it can be useful to think of chord functions not as behaviors,
but as positions. that is, if you have a certain cadential structure,
the slots within that structure will tell you which kinds of chords are allowed to go
there. like, one of the most common structures in
classical music is what’s called the authentic cadence, which is a dominant chord going to
a tonic. now, in this case the dominant is almost always
the V, and the tonic is always the I, but viewing it first in terms of chord functions
instead of actual chords helps to better model their behavior. you can think of it as the difference between
“this chord is the V, so it’s probably dominant” and “this chord is dominant, so it’s probably
the V.” this may seem like a really fiddly distinction, and it is, which is why I usually
don’t bother with it, but thinking of function in terms of position is gonna be super helpful
in understanding Tagg’s approach. when it comes to modern music, though, functions
can get a bit blurry, because composers are no longer following the same rules that Bach
and Mozart did. those clear, functional progressions often
sound corny and dated these days, so artists have found new ways to drive their songs forward. for example, we’ve seen a significant rise
in the use of the plagal cadence, which is the resolution from IV to I. the IV chord
is subdominant, which means that, classically speaking, it doesn’t really point back to
I, but it still kinda works as almost like a V-I resolution in reverse. it’s weaker than an authentic cadence, but
that weakness is actually a kind of strength: it doesn’t give you the same clear sense of
finality, so you can do it over and over without creating awkward stopping points, making it
perfect for chord loops which, oh yeah, isn’t that what this video’s supposed to be about? alright, so let’s take a look at some classic
loops. probably the most famous is the Doo-Wop Changes:
(bang) and this follows our classical rules nicely: these first two chords are tonic,
the IV is subdominant, and the V is dominant. this is exactly what we’d expect to see, but
if we take this structure and apply it to another popular loop, the Four-Chord progression:
(bang) things get complicated. suddenly we have the V chord in a tonic position,
the VI is subdominant, and the IV is dominant? that’s… not right. now, we can just about rearrange these functions
into a new model, one that still makes a fair amount of sense, especially if we’re cool
with plagal cadences, but the cracks in this approach are starting to show, and by the
time we get to a loop like the Pixies’ Where Is My Mind (bang) or Radiohead’s Creep (bang)
it becomes really hard to label these chords with functions that even remotely describe
what they’re actually doing, and Tagg argues that that’s because we’ve missed the point. we’re treating these loops as if they’re trying
to lead us somewhere, but they’re not: they’re already exactly where they want to be. so how do we model that? well, first things first, we’re gonna need
some new functions. Tagg labels the four parts of the loop as
the tonic, outgoing, medial, and incoming chords. but naming things isn’t the same as analyzing
them, so what do these functions do? well, basically, this is a fancier version
of the shuttle idea from earlier: the tonic and medial chords are the two poles of our
harmonic space, while the incoming and outgoing chords serve to set them up. but how do you set up a chord? there’s a couple ways. the first one is stepwise motion: you take
a chord, then slide it up or down a little bit to create a sense of arrival. we can see this, for instance, in Creep (bang)
where the outgoing B major slides up to the medial C major. the other main approach is dominant motion,
where we move up or down by a perfect 5th. 5ths are very strong intervals, and we’ve
been trained to hear them as resolutions: that’s the foundation for both the authentic
cadence and the plagal cadence. we can see this again in Creep, where the outgoing C
minor does a plagal resolution to the tonic G, and it also pops up in Where Is My Mind
(bang) for both the outgoing and incoming approaches. of course, not every loop has these obvious
set-ups. in the Doo-Wop Changes (bang) we see that while the incoming chord sets up
the tonic, the outgoing chord doesn’t really set up the medial chord. it’s moving by third,
which is relatively weak and doesn’t give us much of a sense of landing. in fact, instead of being set up by the outgoing
chord, the medial chord is actually setting up the incoming chord by step, which means
it’s not really serving as an opposite pole at all: in this progression, medial function
is subverted in order to more strongly emphasize the tonic. this isn’t a two-root system, which
helps explain why traditional functional harmony does such a good job modeling it. in fact, from my own observations, it seems
pretty common for the medial chord to set up the incoming one, even when it’s also getting
set up by the outgoing chord. I think it’s a way of differentiating it from
the tonic, creating a real root, which doesn’t go anywhere, and then a secondary passing
root, still stable but also subservient. we see this in Where Is My Mind (bang) where
the medial G# slides up a half-step to the incoming A, and it also pops up in one of
the most underappreciated four-chord loops in modern rock, the plagal cascade (bang)
heard in songs like Linkin Park’s What I’ve Done, Green Day’s Boulevard Of Broken Dreams,
and Imagine Dragons’ Radioactive, among many others. it’s basically just a series of plagal resolutions
with each chord setting up the next, except for the tonic, which moves to the outgoing
chord by third, thus establishing itself as the end of the harmonic road. but even fairly simple progressions can make
a lot more sense when viewed through this lens. Tagg cites The Troggs’ Wild Thing (bang) where
traditional functional harmony would tell us that this dominant V chord is being subverted
by an interrupting subdominant IV, but the loop model tells a very different story. here, the V chord isn’t really dominant at
all, it’s just serving as a secondary musical location, which we transition into and out
of via the IV chord. in this case, the tonic is still stronger, so we hear the whole thing
in a single key, but the V isn’t a directional sound, it’s just another place to be. this progression also highlights another important
aspect of Tagg’s theory: if a chord is doing more than one job, it’s more important to
the sound of the progression. in Wild Thing, the D chord is both incoming and outgoing,
which means it’s controlling all the movement of the loop. on the other hand, Sweet Home
Alabama (bang) has G as both its medial and incoming chords, strengthening it so that
it takes on an almost equivalent level of tonal weight to the tonic D, and if you’ve
been around music theory circles or just watched the video I did a couple weeks ago, you’ll
know that there’s a lot of ambiguity about which key this song is in. Tagg says it’s in D, because that’s in the
tonic position, but the fact that G does so much of the functional heavy lifting is a
large part of why that’s hard to nail down. however, Tagg also argues against the idea
that the first chord in the loop must necessarily be the tonic. here, he cites the Beatles’
With A Little Help From My Friends (bang) where we see two consecutive plagal resolutions
setting up a return to the I chord in the third position. here, the loop is perhaps better understood
as starting on the medial and then working its way back to the tonic. there’s also the chorus loop from Hurt (bang)
which starts on the outgoing chord, saving the tonic for last. we’re dealing with loops here, going round
and round in a big harmonic circle, so it kinda doesn’t matter where we start. the chords will always fall in the same order. so if it’s not always the first chord, how
do we know which one the tonic is? well, that all comes down to listening, but to help us
out Tagg offers five things to look for. first, is there another progression elsewhere
in the song that might help us contextualize what we’re hearing? second, does the song end on a specific point
in the loop in a way that sounds like a satisfying resting place? third, are there any rhythmic
or melodic clues in the way the loop is played that help set up one chord more than the others? fourth, is there a chord with more than one
function in the loop, especially if it’s the first or last one played? and fifth, does
the song come from a tradition in which some tonal configurations are just more common
than others? none of these will, on their own, guarantee
that you’ve found the correct tonic chord, but honestly, if you really can’t figure it
out, there’s always another option: maybe there is no single tonic. much like in Great Gig In The Sky, the tonic
and medial chords may just be so similar in weight that determining which one is which
is impossible and, more importantly, unhelpful. ultimately, I think the point of Tagg’s model
is similar to an argument I made in my initial video: a lot of loops aren’t really best understood
as belonging to a larger tonality. they’re their own self-contained thing, governed
less by functional obligations and more by a series of transitions. when you’re looking
at a loop, the best question you can ask yourself isn’t what key you’re in, it’s how the different
chords are setting each other up. of course, just like any other theoretical model, Tagg’s
loop functions shouldn’t be treated as a template: there’s no one right way to write a chord
loop, and you’ll always find exceptions. but it’s a much more flexible model than functional
harmony, and it does a really good job explaining a pretty wide variety of songs. the four-chord loop is one of the staples
of modern harmony, but if you want to turn your loop into a song, you’re gonna need production
skills too, and Skillshare’s got lots of great classes to help with that! I’d especially recommend a course called Inside
The Studio, which they made in collaboration with Grammy-winning producer Focus… he takes
you through his process of collaborating with artists, making and polishing tracks, and
all sorts of other skills a producer needs in order to make their best music. if you want to check that out, Skillshare’s
even offering 2 free months of premium access to the first 500 12tone viewers to click the
link in the description! it comes with complete access to all their amazing course on music
production, songwriting, and lots of different instruments, along with plenty of other skills
a working musician might need like marketing, organization, and design. and if you like what you see, sticking around’s
super affordable, with premium plans starting under 10 bucks a month, so why not give Skillshare
a shot? anyway, thanks for watching, thanks to our
Patreon patrons for making these videos possible, and extra special thanks to this video’s Featured
Patrons, Susan Jones and Jill Sundgaard. if you want to help out, and help us pick
the next song we analyze too, there’s a link to our Patreon on screen now. you can also join our mailing list to find
out about new episodes, like, share, comment, subscribe, and above all, keep on rockin’.

You may also like


  1. The first 500 people who click this link will get 2 free months of Skillshare Premium: https://skl.sh/12tone24

    Some additional thoughts/corrections:

    1) One thing I should note here is that as far as I could find, Tagg never explicitly lays out his model in significant detail: He describes its basic principles and then starts applying it to examples, which is a perfectly valid approach, but it means that I've had to fill in some gaps with my own interpretations and observations. To the best of my knowledge, everything in the video reflects his approach (Except the part about the medial chord setting up the incoming, which, again, is my own addition.) but if you'd like to read it more directly, the book is called Everyday Tonality, and I've included a link to buy it in the sources in the video description.

    2) A point that I left out is that Tagg's definition of loop requires a certain length. Basically, he works from a principle called the "extended present" (Basically, the amount of time you can experience as roughly a single event, which he defines as between 1-8 seconds depending on tempo, style, and other concerns.) and then argues that a shuttle must exist within the extended present and a loop can take up to two of them. This means, for instance, that Creep, which I cited, isn't technically a loop since it takes about 20 seconds to repeat. I'm not really convinced that this is a necessary condition, but it's certainly true that the length of time you spend on each chord does impact perception, so slower pieces feel different from faster ones. I still think the model works even beyond Tagg's 16-second limit, though.

  2. When you look at the Melody the great gig in the sky wouldn't you say that the composition is mostly in g Dorian? It seems more confusing to say that there's 4 possible tonal centers when it seems to melodically come to rest on g pitches. Pink Floyd uses Dorian alot (like breathe on the same album)

  3. Four chord loops are what musicians love because the general public wants shit repetitive music. Just take "despacito". Ugh! All pop music sounds the same.

  4. It's funny how you focused on "Great Gig in the Sky" at the beginning, because just last week Rick Beato posted a video about "Rocket Man" and when he demonstrated the first two chords in the piano part I immediately thought, "Hey, that's 'Great Gig in the Sky'!"

  5. You could look at this video as the main topic being the tonic and the skillshare promo being the subdominant but really it's all about the transition.

  6. From what I gather, musicians love the four chord loop because there’s actually a lot of complex iterations of it… Yet they all tend to resolve well, which is what fans love

  7. Interesting. Really enjoyed the stuff on the chord shuttle. LCD Soundsystem use it a lot in their work (All My Friends, Dance Yrself Clean, Home etc.)

  8. this is amazing, when i heard the plagal cascade it felt amazing i like it so much. (7:03)
    immediately my head went racing with ideas 😛

  9. I once wrote a chord loop that goes Dm Am Cm D7. I really liked the sound of it but I had no idea what key I should say it was in. Now I can rest easy knowing that that's okay 🙂

  10. I went back to the previous video on 4-chord loops and noticed something missed from the discussion about Creep. You point out that the G has D as a 5th and B has D# as a 3rd, but if we continue onwards C has E as a 3rd and Cm has Eb/D# as a 3rd, giving us a nice little D -> D# -> E -> D# motion. I'd argue that the B and Cm chords both work well as outgoing/incoming chords because that D# isn't in the G Major scale and thus doesn't feel as restful as the G or C do.

    (Tagg's idea works really well for Creep)

  11. One thing I like about this model is that makes more a point of the oft-overlooked fact that a chord's place in a progression has a huge impact on how it's perceived. 8-bit Music Theory talks about it a bit in his Sonic video.

  12. ever since i started thinking of chords less as their names and relations and thinking a bit more of them as “baaaah baaah” my improvisation has felt a lot more post-punk and that’s a pretty pretty cool

  13. Another fantastic video!
    However, I disagree with Where Is My Mind being a “true” example of your point.

    In the chord progression that you highlighted, I kind of agree with you. There are moments in the progression that you give and in the song itself that seem to suggest a bouncing between two keys: the tonic (E) and the relative minor (c#). (WHERE is my key center??) The primary progression that you use in the video (E c#m G# A) is a great example of 2 “locations” we’re bouncing between, where the G# serves as a V to take us to the bVI (A) of the new key c#m, while the A also serves as a subdominant moving back to the E.

    However, that’s not the only progression to the song. The following section briefly gives us an indication that we’re now in the key of c#m with (E G# A am c#m). It’s then immediately followed by a big, resounding B7 for a couple bars, which is the V7 that takes us back to our original key of E. This is where I believe the song destroys true ambiguity. While it begins with a strong tonic/subdominant/minor-modulation “bouncing between locations” feeling, its huge resolution is ultimately V-I. (An “answer” to our wobbly confusion). To me, this changes the way I hear the song: that it’s truly in the key of E and these “moments” of modulation are ultimately brief.

    I see and hear too much direction to truly classify the song itself as an example of ambiguous chord function. I don’t think you’re “wrong”, (I mean, c’mon in theory there is no such thing) but I do think there are more clear examples out there where the song itself demonstrates what you’re talking about; never really giving us an “answer” as to how the chords are functioning or what is its definite key. This ambiguity is why Loopers love these progressions. It gives them a tiny bit of structure, but ultimately allows them to play around with the tonality/feeling, and it also allows them to incorporate many color tones in their melodies.

    This was a great video and I would love to see you explore this concept more in future videos, especially since most pop songs of today dwell in this ambiguous 4-chord territory. I feel like Nirvana would be a great band to use as well, because many of their songs use 4 chord repetitions, none of them actually containing a true “function”.

    Thanks again, man. 🙂

  14. First chord doesn’t have to be I (tonic) is such a tough thing to get my students to grasp.

    Look at Daft Punk’s “get lucky”

    Bm, Dmaj7, F#m7, E

    iv, VI, i, VII

    At least, that’s how I hear it. Anyone else got a different approach to it?

  15. The beginning of this video reminds me a lot of 8-Bit Music theory's video on how Chrono Trigger uses non-functional harmony.

  16. Idk.. all I heard from all this is "Because it sounds good" but with extra steps to make it sound more theoretical

  17. In my musical journey,, I first learned the musical therory by myself by analysing the songs I loved. My main influences were pop and rock and I didn't really care about classic music. I was seeing each chord of a chord progression as a colour and a world by itself instead of a functional tool. I didn't even know the concept of dominant so it didn't bother me when the V chord didn't lead to the I but I loved the sound of that chord it was one of my favorite colour. One of my favorite song was Hotel California and the chord progression begins with | i | V | bVII |. The melody just make me feel that we were visiting a different colour at each chord but I wasn't (or maybe uncounsciously) feeling the direction of the chord. Then I go to the university to learn Musicology and I find it really easy because I knew a lot of thing even if I was naming and feeling them. After my Licence my perception changed and I feel more the functional harmony. And I think it's the case of most of the people who are just listening to modern music.

    Damn it's way easier to understand then to explain in english, I'm french..

  18. "Much like in Great Gig in the Sky, the tonic and medial chords may just be so similar in weight that determining which one is which is impossible,"… and who is who 🎵

  19. Yeah, o.k, I just hope this doesn't lead to brain damage. I composed a 4 chord loop ones, little did I know it was good enough. I will dig in to my old tapes one day and think of this video.

  20. Hey! Listen! 4 disgruntled fairies disliked because the drawing for "unhelpful" 😂. I absolutely love your videos so I'll look the other way. ✨

  21. I didn't really think about how bizaare and difficult to interpret the pictures he keeps drawing are until just now.

  22. Although you mostly do more popular songs, I'd really like you to analyze one of Igorrr's songs ("Opus Brain" would probably be a good candidate) since the mix of baroque music, electronic music and death metal is both very unique and harmonically interesting.

  23. I think it's arguable that in a doo wop progression that the outgoing chord does set up the mediant in Tagg's analysis.

    Even if there's nothing particularly inherent in the note motion to directly point from a Dm to a Bb (such as the example chords given in the vid), the echoing of the motion from the chord transition beforehand (F to Dm, all notes dropping a triad) establishes the direction of movement sufficiently enough to prepare listeners to feel that the Dm>Bb (all notes dropping a triad again) should be an expected progression. It effectively requires the effort of both the tonic's and the outgoing's movement to get to mediant; a direct continuation of the shaky path already started rather than a smoother detour to get to the same location. The weakness of the resolution from the outbound vi chord to the mediant IV chord is balanced by the way that the movement then rebounds back upwards from iv to V, preventing a complete loss of strength in the mediant chord.

  24. Those four people that disliked this video are likely guilty of abusing the four chord loop 😝 I look forward to all of your videos ❤️

  25. I feel like a lot of the people complaining about modern four chord loops are the people who play three chord loop blues

  26. I love it when you cover non-functional harmony this practically, because things like this are also the key to writing in the full range of modes.

  27. When he said it could be in four possible keys I immediately though “G minor, C, or F.” Proud of myself. I’m learning!

  28. you make music sound more complicated than the most complicated of nuclear physics..so many meaningless words, so hard to listen to

  29. My three favourite four chord songs are these 🙂
    – Killed By Death (Motörhead)
    – Musik ist keine Lösung (Alligatoah)
    – Robeast (Dance With The Dead)

    The third one of these in particular got an epic signature part around the one minute mark that does not start on the tonic. It starts on the flat minor sixth. And with the first minute of the song building up to that, the tension works great. That's probably DwtD's favourite chord loop. Other great tracks of them with the catchiest part also starting on that flat minor sixth would be their Remix of Scandroids' "Neo Tokyo", as well as their original "The Man Who Made A Monster" 🙂

  30. man I don't understand how you communicate so many ideas through visual, sound and language. If I can absorb 5% of this, I'm happy.

  31. I got a progression like that thing, chord shuttle.
    The verse is B E and the Pre-Chorus would be A B C#m D#m7 C#m7 B A
    And then resolves by playing A E

    But D#m isn't in E, and A isn't in B either.

    Yeah, I'm pretty proud of it.

Leave a Reply

Your email address will not be published. Required fields are marked *