Music Hub › Guides › Theory and harmony › Chord substitutions and reharmonization, explained in depth

The principle behind every reharmonization: substituting without breaking function

Reharmonizing a progression means replacing one or more of its chords with others, without the song losing what makes it recognizable: its form, its melody, the place where it breathes. This guide assumes that seventh-chord construction and the ii-V-I cadence, jazz's most frequent progression, are already familiar ground: they're covered in depth elsewhere on the Music Hub and won't be repeated here. What concerns us here is different: how to replace one chord with another without the ear losing the thread, and why some replacements work while others simply sound wrong.

The answer almost always comes down to the same mechanism: two chords that share several notes substitute for each other without a jolt, because part of the sonic material stays identical while another part changes. It was functional theory, popularized from 1893 onward by the German theorist Hugo Riemann in his treatise Vereinfachte Harmonielehre, that best formalized this principle by grouping a key's seven degrees into three functional families: tonic chords, carried by the first, third, and sixth degrees; subdominant chords, carried by the second and fourth degrees; and dominant chords, carried by the fifth and seventh degrees. Within a single family, two chords generally share two notes out of three, which explains why they can be swapped without distorting the function of the passage.

This principle of common tones isn't limited to diatonic chords within a single key. It also structures tritone substitution, which swaps two dominant chords sharing the same tritone, the passing diminished chord, which is often just a disguised secondary dominant, and modal interchange, where only one note of the chord changes color while the other two stay put. Seven techniques are covered in this guide; all of them exploit the same logic, at different degrees of distance from the starting key. The table below summarizes them before each section goes into the details, notes included.

One remark before getting into the details: substituting a chord almost always changes the bass note being heard, and often part of the tensions perceived by the other musicians playing the same chart. The principle of common tones explains why a substitution works in theory; it doesn't excuse skipping the practical check that it also works with the melody and with the rest of the band, a point covered in the final section of this guide.

Diatonic substitution by third: I, iii, and vi, and the neighboring families

The tonic family illustrates the principle of common tones most simply. In C major, the first-degree chord is C major, C-E-G. The sixth-degree chord, A minor, stacks A-C-E: it shares two notes out of three with the tonic chord, C and E, and differs only in its root, A instead of C. The third-degree chord, E minor, stacks E-G-B: it too shares two notes with the tonic, E and G. Replacing a C major with an A minor or an E minor therefore doesn't upend the function of the passage, the ear still recognizes the essential harmonic color, but it noticeably darkens the result, since both substitute chords are minor.

The same principle applies to the two other functional families, with one fewer common tone each time. The subdominant family groups the second and fourth degrees: D minor, D-F-A, shares F and A with F major, F-A-C. The dominant family groups the fifth and seventh degrees: G major, G-B-D, shares B and D with the diminished chord on the seventh degree, B-D-F. A ii chord can therefore replace a IV in a plagal cadence without changing its meaning, just as the diminished vii°, rarely played on its own given how unstable it is, can appear in place of a V in a context calling for a more fragile, more tense dominant.

The third-degree chord deserves a caveat that Riemann's own theory already pointed out: E minor doesn't just share two notes with the tonic C major, it also shares two with the dominant G7, G and B. Riemann in fact classified the iii in both families at once, tonic and dominant, unable to settle on one or the other. In practice, this ambiguity makes the iii less reliable than the vi as a tonic substitute: depending on the melodic context and the bass motion around it, it can sound like a languid tonic or, just as easily, like a dominant that hasn't resolved yet. The vi remains the safer choice when the goal is to darken a tonic without any ambiguity about its function.

In a chart, this family of substitutions mainly serves to let a tonic chord breathe when it would otherwise return too often unchanged: replacing the second occurrence of a C major with an A minor at the same spot in two consecutive phrases introduces a new color without the listener losing the sense of still being on the starting chord. It's one of the most discreet reharmonization tools around, precisely because it never shifts the function, only the color.

Tritone substitution: when the third and the seventh swap roles

Tritone substitution replaces a dominant chord with another dominant chord, built a tritone away, either six half steps up or down: the distance comes out the same either way since the tritone splits the octave into two equal halves. In C major, the dominant chord G7, G-B-D-F, can be replaced by Db7, built a tritone above it.

The proof by common tones is straightforward. G7 contains a tritone between its third, B, and its seventh, F: it's that tritone that carries all the chord's tension and pulls toward resolution. Db7 stacks Db, F, Ab, and B, spelled Cb in strict notation but identical to B by ear: its third is F, its seventh is B. The two notes of the tritone haven't changed at all, only their role has flipped, the third of one becoming the seventh of the other and vice versa. It's this exact swap, third for seventh, that lets Db7 resolve to C major exactly the way G7 did, with the same feeling of release.

What does change, on the other hand, is the bass motion. The usual cadence has the root descend a fifth, from D to G and then from G to C. Tritone substitution replaces that descending fifth with a chromatic walk, from D to Db and then from Db to C, each step now only a half step. It's this chromatic bass line, more unusual and more colorful to the ear than the fifth leap, that made tritone substitution one of bebop's tools of choice in the 1940s. The tritone, already central to classical harmonic language long before jazz, became the style's most characteristic interval there, to the point that some historians of the genre describe it as bebop's signature interval, used just as much in chord charts as in the melodic lines of musicians like Charlie Parker or Dizzy Gillespie.

For improvising, a substituted dominant chord strictly calls for a different scale than the original V7's, the lydian dominant scale, which alters two notes of the reference major scale: the fourth, raised a half step, and the seventh, lowered a half step. That topic goes beyond the scope of this technical overview: the main thing to remember is that tritone substitution changes the bass note heard by the other musicians, which makes it a tool to announce before playing it rather than one to improvise alone in the middle of a band still following the original chart.

The secondary V7: briefly tonicizing any degree

The principle behind a secondary dominant is to borrow, for the length of a single chord, the leading tone of a neighboring key in order to briefly tonicize a degree of the chart that isn't the tonic. The technique is nothing specific to jazz: it's already found systematized in the music of Johann Sebastian Bach, nearly two centuries before functional theory gave it its current name, and it has remained ever since one of the most ordinary tools in the whole of Western tonal harmony.

The mechanism is always the same: to tonicize a chord X, you build its own dominant, as if X were momentarily the tonic of a neighboring key, by altering the one note needed to give it the major third and the tritone of a genuine dominant chord. In C major, the second degree, D minor, can thus be preceded by A7 rather than its natural diatonic chord, A minor: the third of the diatonic A minor, C, is raised a half step to C♯, which gives A its leading tone and turns the chord into a genuine dominant, ready to resolve to D. This borrowed dominant is notated V7/ii, read as 'the V of ii,' a convention used throughout this guide.

Each of these secondary dominants alters only a single note of the diatonic chord it replaces or precedes, which explains why they fit so easily into a chart without giving the impression of a key change: the ear perceives a passing chromatic color, never a real modulation. The most commonly used of the four is likely the V7/V, D7 in C major, which prepares the arrival on the dominant chord with the same energy as a miniature ii-V, and which turns up identically in plenty of jazz standards as well as pop songs, often without the casual listener even noticing it.

The secondary V7 shouldn't be confused with a key change: the chart stays entirely within the starting key, only a single note takes on a chromatic color for the duration of one chord before falling back in line. It's this economy, just one degree altered for a clearly audible effect of tension, that makes the secondary dominant as discreet to work into a chart as it is powerful to the ear.

The inserted ii-V: preparing a target chord with its own ii-V

A secondary dominant prepares its target degree with a single chord. The inserted ii-V goes further: it prepares that same target degree with its own complete ii-V cadence, exactly as if the destination chord became, for the space of two bars, the tonic of a miniature key. The technique doesn't replace anything already in the chart, it adds an extra chord in front of a secondary dominant that may already have been sitting there.

Let's return to the example of the second degree in C major, D minor, prepared by its secondary dominant A7. To build the inserted ii-V that precedes it, all it takes is treating D minor as the tonic of a miniature minor key and applying to it the minor ii-V cadence covered in the guide on the ii-V-I cadence: the second degree of D minor is E half-diminished, its fifth degree, altered to recover its leading tone, is A7. The insertion then gives E half-diminished, then A7, then D minor 7, or Em7♭5-A7-Dm7, where the chart previously held only a single A7 chord, or even nothing at all before arriving on D minor.

The same principle applies to a major target degree, with one difference in color on the inserted ii: if the destination chord is major, like the fourth degree F major, its own second degree is a plain minor chord rather than a half-diminished one, since we're then dealing with the major ii-V-I cadence rather than its minor version. Preparing F major with its own ii-V gives G minor, then C7, then F major, or Gm7-C7-Fmaj7, a motion that locally recreates, right inside the key of C major itself, the exact structure of a complete cadence toward another tonic.

The I-vi-ii-V turnaround, already covered in the guide on the ii-V-I cadence, lends itself particularly well to this treatment: each of its four chords can receive its own inserted ii-V in front of it, turning four bars into eight without ever leaving the starting key, only slowing down the perceived harmonic rhythm and multiplying the micro-resolutions. It's a technique characteristic of club jazz, where a familiar standard gets progressively reharmonized on every chorus, each new insertion simply requiring you to identify whether the target degree is major or minor before applying the right form of cadence to it.

The passing diminished chord: a chromatic chord hiding a secondary dominant

The passing diminished chord connects two diatonic chords through chromatic motion, slotting in between them a half step below the second one. Between the tonic chord C major and the second-degree chord D minor, for instance, you can insert C♯ diminished 7, built a half step above C and a half step below D.

What makes this passage convincing to the ear isn't just the chromatic bass motion, it's also a hidden harmonic identity that the notes themselves reveal directly once you work them out. C♯ diminished 7 stacks C♯, E, G, and B♭, four notes each separated by a minor third, the symmetrical structure typical of every diminished seventh chord. Now take A7 with a flat ninth, the secondary dominant of the second degree covered in the previous section: it stacks A, C♯, E, G, and B♭. Remove the root alone, A, and what's left is C♯, E, G, and B♭, exactly the four notes of the diminished chord. A passing diminished chord is therefore, most of the time, nothing but an altered secondary dominant played without its root, a reading that explains both why it pulls so hard toward the following chord and why its use stays so close to that of the secondary V7 covered above.

This kinship has a useful practical consequence: since a diminished seventh chord is perfectly symmetrical, stacked in identical minor thirds from bottom to top, each of its four notes can be read as the flat ninth of a different dominant, a major third below it. C♯ diminished 7 therefore works equally well as a fragment of A7♭9, C7♭9, E♭7♭9, or F♯7♭9, which could just as easily be spelled G♭7♭9, the enharmonic ambiguity being precisely the nature of this chord: four different readings for one and the same shape played on keyboard or guitar, which explains this chord's popularity among musicians looking for a chromatic passing tone without having to remember four different fingerings.

The technique goes back at least to stride, the piano style born in Harlem in the 1920s around musicians such as James P. Johnson, who routinely approached a diatonic chord with a diminished seventh chord built a half step below it. This usage then spread through the language of swing and later bebop, where the passing diminished chord remains today one of the most commonly used tools for connecting two neighboring chords without breaking the chromatic motion of the bass line.

Modal interchange: bVI, bVII, and the minor iv borrowed from the parallel mode

Modal interchange (also called modal borrowing) means replacing a diatonic chord with its equivalent built on the notes of the parallel minor mode, the one sharing the same tonic but not the same key signature. In C major, the parallel minor mode is C minor, whose key signature lowers the third, sixth, and seventh degrees relative to C major: E becomes E♭, A becomes A♭, B becomes B♭.

The fourth degree illustrates the mechanism most simply. Diatonic, F major stacks F-A-C. Borrowed from the parallel minor mode, it becomes F minor, F-A♭-C: two notes out of three stay exactly the same, F and C, only the third drops a half step, from A to A♭. This minimal change, just one note shifted by a half step, is enough to completely darken the chord's color without touching its root or its subdominant function.

The borrowed sixth degree follows the same logic one half step further out: instead of A minor, the natural diatonic chord, you borrow the lowered sixth degree of the parallel minor mode, A♭ major, A♭-C-E♭. This chord shares two notes with the borrowed iv minor just discussed, A♭ and C, which explains why the two so often travel together in the same passage, one naturally leading into the other. The borrowed seventh degree follows the same principle one step further still: B♭ major, B♭-D-F, in place of the diatonic vii°.

This last borrowed chord, B♭ major 7, enables an alternative cadence well known in jazz as the backdoor progression: minor iv, then bVII7, then I, or F minor 7, B♭7, C major. Its resolution works through a precise voice-leading motion rather than the tension of a classic leading tone: the minor seventh of the B♭7 chord, A♭, drops a simple half step to G, the fifth of the C major tonic chord. This is a plagal-type resolution motion, inherited from the same borrowing that gives the minor iv its color, rather than the leading-tone motion proper to a genuine V7. The result sounds softer, less conclusive than a classic ii-V-I cadence, which makes it a favored alternative at the end of a verse or section, wherever a less emphatic resolution is called for.

Unlike the other substitutions in this guide, modal interchange doesn't try to preserve a maximum of common tones with the chord it replaces: on the contrary, it's the change of a single specific note, almost always the third, that carries the whole effect. The trade-off risk is giving the impression of a key change when it's really just a momentary borrowed color: the chart stays in C major from start to finish, only a few chords borrow, for a bar or two, from its parallel minor mode.

Pedal point: a fixed note under a changing harmony

Pedal point flips the principle of the earlier substitutions around: instead of replacing one chord with another, it keeps a single bass note fixed while the chords that would normally change above it keep succeeding one another. The term comes directly from the organ: in baroque music, and especially in Johann Sebastian Bach's organ fugues, the organist holds a low note on the pedalboard, played with the feet, while the hands carry on their harmonic discourse on the manual keyboards, hence the French name point d'orgue also given to this device.

Technically, pedal point forces a single common note onto an entire succession of chords that don't necessarily belong to that same harmony. Take the ii-V-I cadence in C major again: instead of letting the bass descend from D to G and then to C, you can hold G under all three chords, D minor, G7, and C major. D minor set over a G bass is then reheard as a form of G eleven, its own notes F and A forming the minor seventh and the ninth of that implied G respectively; G7 finds its natural root again, unchanged; C major set over that same G bass is reheard as C over G, a suspended color that delays the feeling of full arrival until the bass itself finally reaches C.

Two uses dominate in practice. The dominant pedal, as in the example above, holds the dominant note under a succession of chords leading toward the tonic: it's used almost always to announce an approaching resolution, particularly at the end of a chorus or right before a concluding cadence, where it builds up tension before releasing it all at once when the bass finally reaches the tonic. The tonic pedal, conversely, holds the tonic note under chords that, above it, change freely, including chords that don't belong to its own harmony: this is the most frequent use at the opening of a tune or during an instrumental bridge, where it gives a stable anchor while the harmony explores above it without ever really leaving home.

Pedal point isn't, strictly speaking, a chord substitution: it's an addition, a constraint placed on the bass that reinterprets the function of everything played above it. It's precisely this absence of change, the single note that refuses to move while everything else evolves, that creates the tension characteristic of this device, a tension all the more pronounced the further the held note strays from the function of the chord played over it.

Reharmonizing: when it enriches a song, and when it destroys it

Every technique in this guide shares the same risk: it works in theory, common tones and all, but guarantees nothing about the musical result once played against a melody, a singer, or a whole band. Whether reharmonizing enriches a song or destroys it isn't a question theory alone can settle.

A reharmonization most often enriches an instrumental piece or an arrangement worked out in advance, where every musician knows the new chords before playing: a solo piano intro or ending, an extra chorus in a jazz standard progressively reharmonized on each pass, a vocal arrangement where the harmony is built specifically around the lyrics. In these contexts, the substitution adds a color the listener perceives as an interesting variation on material they already know, precisely because the melody and the form stay stable while the harmony renews itself.

It destroys a song more often than people think, though, for three distinct reasons. The first is melodic: every substitution changes the notes of the chord, and a melody note that sounded like a consonant third or fifth over the original chord can become an unprepared dissonance over the substitute chord. Before reharmonizing a passage, it's essential to systematically check that every melody note still works over the new chord, either as a chord tone or as a clearly intentional tension, otherwise the substitution sounds like a mistake rather than a choice.

The second reason has to do with recognition: a familiar song draws a good deal of its identity from its chord progression as much as from its melody, especially for an audience that sings or plays it themselves. Stacking several substitutions into the same chart, even if each one is technically justified, can make a well-known tune unrecognizable, which shifts the listener's attention from the song to the performance of whoever is reharmonizing it, rarely the intended effect outside a context where that's precisely what people came to hear.

The third reason is stylistic: certain styles, folk, punk, a good deal of traditional rock, draw their strength from the simplicity and predictability of their harmony, and a clever reharmonization sounds out of place there rather than enriching, no matter how refined its theory. The right instinct, before any substitution, is to ask whether the song's style, the audience listening to it, and the musicians accompanying it expect or tolerate that kind of color, and to save the boldest substitutions, like tritone substitution and the passing diminished chord stacked on top of each other, for contexts where everyone is certain to follow the same change at the same instant.

Common pitfalls to avoid

Explore in the Music Hub

More guides