Music Hub › Guides › Scales and improvisation › Building a Solo: Phrase, Motifs and Silence

What the scales already know, what this guide adds

This hub's collection already includes three guides entirely devoted to note choice in improvisation. One details the major pentatonic scale and its relative minor, the way they're built by removing the two half-step-carrying degrees from their parent scale. Another details the dorian and mixolydian modes, the two most commonly used over a static chord or vamp. A third completes the picture with the lydian, phrygian, aeolian, and locrian modes. A fourth guide, on the ii-V-I cadence, explains in detail which scale to match to each chord of a moving progression: dorian over the ii, mixolydian over the V, major scale over the I. Together, these four guides answer, in depth, a single question: which notes are you allowed to play, at any given moment, over this or that chord.

None of the four answers a different question, one just as decisive for the quality of an improvisation: how to play those notes once you know which ones to choose. A musician who knows the five positions of the pentatonic scale by heart, the seven modes of the major scale, and the ii-V-I cadence in all its forms can still produce a solo that sounds like a recited exercise, a scale run up and back down, technically correct and yet empty of everything that makes an improvisation tell a story. The difference between the two comes down to no extra note whatsoever: it rests entirely on how the already-chosen notes are organized in time.

This guide assumes that note choice is settled elsewhere in this collection, and doesn't revisit any of the four guides that detail it. It focuses exclusively on the gesture that turns a string of correct notes into a musical discourse: the motif and its repetition, call and response, the role of silence, the placement of chord tones on strong beats, tension and release seen from rhythm rather than harmony, the construction of a chorus that progresses from start to finish, and how to start and then end a phrase. These are gestures, not notes, and that's precisely what this hub's chord charts, however precise they are about which chords to play, never tell you.

The motif: the smallest idea worth repeating

A motif is a short melodic and rhythmic fragment, usually two to five notes, brief enough to be memorized instantly by the ear and identifiable enough to be recognized whenever it returns, even in transformed shape. It's the smallest unit of meaning in an improvisation, the equivalent of a word or a short expression in spoken discourse: a single word tells no story on its own, but a discourse that never strings the same word together twice in recognizable forms doesn't build a sentence either, it just lists things.

The most common confusion for those starting out with improvisation is treating a motif as a simple synonym for a lick, that ready-made fragment worked out note for note in a scale position and pulled out identical from one solo to the next, over whatever chord it more or less fits. A memorized lick can very well serve as the starting material for a motif, but it only becomes one, in the full sense of the word, once it's reworked live, shortened, shifted, repeated with a variation, inside the very solo being played. A motif lives inside the solo that gives birth to it; a lick, on the other hand, already exists before the solo begins and doesn't depend on it.

What distinguishes a motif worthy of the name isn't its complexity, then, almost always the opposite: the shorter and simpler it is, the better it lends itself to being repeated several times without wearing out the ear, and the more room it leaves for the transformations detailed in the next section. A two-note motif, a third interval repeated at different pitches for instance, can carry an entire solo as effectively as a seven- or eight-note shape, provided it's recognizable enough for the listener to retain it from the very first occurrence. It's this economy that makes a motif usable in real time, at the speed of improvisation, whereas an idea that's too long or too loaded becomes hard to rework without replaying it identically, which reduces it back to the status of a mere lick.

Spotting a good candidate motif while improvising has less to do with a conscious decision than with listening to what you just played: the first phrase of a solo, even chosen without thinking, almost always contains a fragment worth hearing again. The next section details how to bring it back without ever repeating it exactly the same way twice in a row, which is what separates a motif that builds a discourse from a tic that merely recurs.

Developing a motif: repeat, vary, transform

A motif that never returns has built nothing; a motif that always returns identical to itself eventually wears out its welcome, exactly like a sentence repeated word for word in conversation. Between these two extremes lies motivic development, the set of techniques that let a motif evolve while keeping enough of its original shape for the ear to keep recognizing it. The Austrian composer and theorist Arnold Schoenberg theorized this principle across writings spanning decades, from a 1917 manuscript that stayed unpublished for a long time to his pedagogical treatise Fundamentals of Musical Composition, written for the most part between 1937 and 1948. He gave it a name that has since become a benchmark in music analysis, developing variation, entwickelnde Variation in German: the succession of forms drawn from a single base motif through a series of variations, a process he compared directly to organic growth. For Schoenberg, this developing variation isn't just one technique among others, but one of the most decisive principles of composition since the mid-eighteenth century, long before jazz turned it into a tool for real-time improvisation rather than composition on paper.

The concrete operations that enable this variation are few and combine freely with one another: replaying the motif at a different pitch without changing its shape, transposition; chaining it several times in a row while transposing it by the same interval each time, sequence, which creates a clear directional movement, often used to build tension; stretching it into longer rhythmic values, augmentation, or conversely tightening it into shorter values, diminution; flipping the direction of its intervals, so what went up now comes down, inversion; or picking up only a fragment of it, often its head or its tail, fragmentation, which relaunches the idea without ever repeating it in full.

None of these operations requires a complex theoretical justification to be heard: that's precisely their strength. A transposition is immediately perceived as the same shape somewhere else; a rhythmic augmentation is perceived as the discourse slowing down after a denser passage; a fragmentation is perceived as a callback rather than a repeat. The listener needs no vocabulary at all to feel that a motif is transforming before their ears, only enough repetition to have memorized it the first time.

The most often cited example of this principle applied to improvisation, rather than to written composition, dates back to 1958: the composer and theorist Gunther Schuller published an essay in The Jazz Review titled Sonny Rollins and the Challenge of Thematic Improvisation, in which he analyzes, measure by measure, an improvisation recorded two years earlier by saxophonist Sonny Rollins, tracking motivic cells picked up and transformed throughout the solo rather than a mere succession of scales. This essay is regarded today as one of the founding texts of jazz analysis, to the point of having launched an entire field of study treating the solo as structured discourse. It also sparked a debate that remains unsettled: several scholars have since questioned whether Rollins built that motivic development as deliberately as Schuller's analysis suggests, a question that stays open but changes nothing essential for anyone improvising today, namely that the ear perceives and values that coherence, whether it's entirely calculated or partly instinctive.

Call and response: talking to yourself and answering back

The call and response principle designates a musical structure where a first phrase, the call, is followed by a second one that answers it, in echo, in contrast, or in complement. This principle doesn't come from jazz, nor even from the blues: it descends directly from West African musical traditions, where it structured communal singing and collective storytelling long before the Atlantic slave trade. Carried to the Americas, it lived on in the cotton and sugarcane fields of the American South in the form of work songs, which set the pace for the collective labor of enslaved men and women, a leader calling out a short phrase that the rest of the group answered in chorus to keep the rhythm of the physical task. That same structure went on to run through the spirituals, then the blues, then gospel, before becoming one of the backbones of jazz and, more broadly, of most twentieth-century African American popular music.

In a group, call and response is heard most directly in the dialogue between two musicians, a soloist launching a phrase and a second one picking it up or commenting on it, or between a soloist and the rest of the ensemble, when the rhythm section answers what the soloist just played with an accent or a motif. But the principle works just as well inside a single solo played by a single musician: nothing stops you from improvising a call and then answering it yourself, two bars later, with a phrase that completes, contradicts, or extends the first one. In a sense, this is call and response applied to the motivic development detailed above: the initial motif serves as the call, its immediate variant serves as the answer, and the alternation between the two organizes the discourse into a series of small dialogues with yourself rather than a single continuous monologue.

This dialogue dimension explains why call and response remains one of the simplest tools to put into practice for anyone starting out with structured improvisation: it requires no additional harmonic knowledge, only the discipline to stop after a short first phrase, leave a silence, the call and the silence are treated further on as two sides of the same gesture, then play a second phrase that consciously positions itself relative to the first, rather than stringing together a third idea unrelated to what came before. A classic teaching exercise, in fact, has two musicians improvise while strictly alternating four bars each, traditionally called trading fours in North American jazz, or eight bars each, trading eights: the constraint of having to answer what the other player just played, rather than unspooling an independent personal idea, forces you to listen and build a real dialogue, a discipline that remains useful solo once you learn to ask yourself the question before answering it.

Silence: the other half of the phrase

In improvisation, silence isn't the absence of music: it's one of the two materials a phrase is built from, on equal footing with the note. A melodic line that never stops cannot, by definition, contain any identifiable phrase, since a phrase is defined precisely by its boundaries, a beginning and an end separated from the rest by a silence or a breath. Removing silence entirely from an improvisation is like removing the punctuation from a written text: the words stay readable one by one, but the overall meaning gets lost in an uninterrupted stream the ear no longer knows where to cut.

Pianist and composer Thelonious Monk left behind, on this exact subject, one of the best-documented sources in all of jazz: in 1960, while a young soprano saxophonist was briefly playing in his group, Monk passed on to him, orally, about twenty pieces of advice on how to play, which that musician, Lacy, wrote down by hand in his own notebook and later made public. The list circulates today under Monk's name, but the hand that wrote it down was Lacy's. Two of those pieces of advice deal directly with silence: one recommends not playing nonstop and letting some ideas pass rather than playing all of them, the other goes even further, stating that what a musician chooses not to play can matter more than what they actually play. Coming from a musician whose style is still recognized, to this day, for its deliberate use of empty space just as much as of the notes themselves, these two pieces of advice aren't a vague aesthetic stance: they describe a concrete building principle, applicable phrase after phrase.

Silence actually fills several distinct functions in an improvisation, worth telling apart rather than lumping under a single word. First, it marks the boundary between one phrase and the next, exactly the way a period separates two written sentences, letting the listener treat the solo as a sequence of ideas rather than a single wall of sound. It also creates anticipation: a silence slightly longer than expected, right before a note you can feel coming, heightens the tension with which that note will land, a mechanism revisited further on in this guide under phrasing tension. Finally, it sets off whatever surrounds it: a dense phrase framed by silence holds attention far better than a phrase of identical density drowned in an uninterrupted stream of notes.

A useful reflex for working silence in without it becoming an awkward void is to treat it as a duration in its own right rather than as a gap: counting the beats of a silence mentally with the same precision you'd count the beats of a held note, rather than just letting it happen for lack of an idea. A silence that's owned, placed consciously, is immediately distinguishable to the ear from a hesitation or a blank, even though, technically, neither contains a single note.

Targeting chord tones on strong beats

Choosing which scale to play over a given chord is a subject of its own, treated in detail elsewhere in this collection, notably in the guide on the ii-V-I cadence: once that choice is made, every note of the chosen scale is, in theory, available at any moment. But they're not all equal at any given instant, and that's precisely what the technique of guide tones and target notes comes to correct. These two terms, largely synonymous in jazz pedagogy, most often refer to a chord's third and seventh: the two notes that, on their own, determine its exact color, major or minor for the third, dominant, major, or minor for the seventh. A root confirms the chord without saying anything about its nature; a third and a seventh, on the other hand, say everything.

The principle boils down to something simple: rather than running through the chosen scale note after note and hoping the result sounds right, you spot the third or seventh of the upcoming chord in advance, and organize the phrase so that one of those two notes lands precisely on a strong beat, usually the first or third beat of a bar in four. The rest of the phrase, the notes leading up to that target and those leaving it, can then afford more freedom, chromatic passing tones, approaches from above or below, without ever hurting the overall harmonic clarity, precisely because the listener hears the note that matters landing in the right place.

This hierarchy among a chord's degrees also explains why two musicians playing rigorously the same scale over the same chord can sound very different: the one who consciously targets their guide tones on strong beats produces a line whose chord changes the listener clearly perceives, even with their eyes closed, while the one who plays the same notes without that hierarchy produces a line that's correct on paper but harmonically flat, as if the accompaniment were changing without the solo really taking any notice. The difference comes down to no forbidden note and no added note: it rests entirely on the rhythmic placement of a handful of notes already available in the chosen scale.

A secondary benefit of this technique, once it becomes a reflex, has to do with voice leading from one chord to the next: the third or seventh targeted on the arriving chord is often, as the guide on the ii-V-I cadence details regarding tritone resolution, only a half step or a whole step away from the note you just left on the previous chord. Targeting your notes on strong beats therefore amounts, almost mechanically, to producing a line that closely tracks the voice movement of the underlying harmony, without needing to calculate it note by note in real time.

Tension and release in phrasing, beyond the chord

Tension and release are usually studied on the harmonic side: that's exactly what the guide on the ii-V-I cadence details regarding the tritone contained in the dominant chord, which pulls the ear toward the tonic chord that follows. That harmonic tension exists independently of the soloist, it's written into the chart itself, and playing over it neither increases nor decreases it. Yet there's a second tension, one that belongs to phrasing rather than to the chord, which note choice alone doesn't explain: the tension created by the rhythm and placement of a phrase in time, regardless of the harmony behind it, including over a perfectly stable chord that doesn't move at all.

This phrasing tension is built through several means that have nothing to do with harmony. Deliberately delaying the arrival of a target note, making it wait one beat longer than the ear anticipates, creates a tension that exists only in the relationship between the phrase and time, not between two chords. Placing a note outside its expected rhythmic position, ahead of or displaced from the nearest strong beat, a device musicians call displacement or syncopation depending on its degree, produces the same kind of anticipation: the ear knows where the note should land and perceives the gap as a tension to resolve, even if the note itself is perfectly consonant with the chord of the moment. A phrase that deliberately stops before its logical ending, that leaves a question without an immediate answer, also creates a purely structural tension, resolved only a few bars later when the answer finally arrives, as discussed above regarding call and response.

What distinguishes these two tensions, harmonic and rhythmic, is that they can exist independently of one another. A phrase can follow the chord to the letter, without a single outside note, and yet create no tension at all if it lands exactly where the ear expects it, at the right tempo, without the slightest displacement. Conversely, a phrase can stay on the most consonant notes of a perfectly stable chord and still sound full of tension, purely because it delays its arrival, shifts its accents, or leaves a question open longer than expected. Good phrasing generally combines both levers rather than relying on just one: the harmonic tension from the guide on the ii-V-I cadence supplies the material, the phrasing tension described here supplies the exact moment that material needs to arrive to produce its maximum effect.

Starting a phrase: on the beat, before, after

The starting point of an improvised phrase weighs on its character just as much as the notes that make it up, and yet it comes down to a simple choice between three rhythmic options. The first, the most direct, is to start exactly on a strong beat, usually the first beat of the bar: the phrase settles in from the outset with clarity, with no ambiguity about where it begins, a choice that suits the opening of a solo or the return of an already-heard motif particularly well, when the priority is for the listener to immediately identify what's starting up again.

The second option is to start before the beat, as an anacrusis, with one or more notes that prepare the arrival of the strong beat rather than coincide with it. This early start gives the phrase momentum, an impression of movement already launched before the bar has even truly begun, an effect found throughout much of popular song as much as in instrumental improvisation, where very few memorable melodies actually start right on the first beat of their first bar.

The third option, rarer and more marked, is to start after the beat, leaving the start of the bar empty before bringing the phrase in with a slight delay. This delayed start produces a feeling of nonchalance or suspension, as if the phrase were hesitating before committing, and works particularly well right after a silence already established by the previous phrase: the extra delay stretches the wait by one more fraction of a second than the silence alone had already created.

None of these three options is inherently superior to the other two: it's alternating between them, over the course of a single solo, that keeps every phrase from sounding alike. A chorus that starts each of its phrases exactly the same way, always right on the beat for instance, ends up sounding mechanical even if the notes change from one phrase to the next, for a reason that comes down purely to rhythm and not at all to melody. Consciously varying these three starting points, keeping in mind which one was used last, is one of the simplest and most effective adjustments for giving a solo the variety that a succession of different notes doesn't always produce on its own.

Ending a phrase: where the final point lands

If the start of a phrase plays out across three rhythmic options, its ending plays out on a choice of a different nature: which note, and how stable, carries the final point. A phrase can close on a stable note, usually the root or the third of the chord at that moment, a choice that clearly closes the idea and leaves the listener with a sense of completion, especially useful at the very end of a chorus or just before another musician takes over.

It can also close on an unstable note, an unresolved seventh, a ninth, or a note foreign to the chord held all the way through without ever resolving down to a consonance, which instead leaves the phrase hanging, a question rather than a statement. This choice works particularly well in the middle of a solo, when the goal isn't to conclude but to redirect attention toward the next phrase: an unstable ending acts like the call in call and response, it almost mechanically demands a follow-up, whereas a stable ending closes the loop and demands nothing further.

The third possibility is to not really end on a note at all, but to let the phrase fade into silence, the last note dying away naturally, or to cut it off abruptly before it reaches its logical conclusion, a deliberate break rather than a conclusion. This last option, rarer, works as a form of deliberate restraint: it suggests there was more to play without ever playing it, which connects directly to the role of silence detailed earlier in this guide, in its most radical form.

The length of the phrase itself also matters in how its ending is perceived. This collection's guide on song structure details why Western popular music organizes its sections into multiples of four bars, a preference tied to the way the ear groups time in pairs; that same preference for two- or four-bar groupings applies, on a finer scale, to the phrases of an improvised solo. A four-bar phrase that closes exactly at the end of the fourth bar sounds natural and expected; a phrase that spills over by a bar or stops a bar too early stands out immediately, for the same reason an irregular song section stands out: the ear has memorized, without even consciously thinking about it, the shape of the previous phrases, and judges every new phrase against that yardstick. This overrun isn't a mistake in itself, exactly as at the level of an entire section: it's an expressive tool, provided it's a choice rather than an accident.

Breathing like a singer: phrasing in units rather than in continuous flow

A singer has no choice: they must breathe, and every breath naturally cuts their performance into phrases whose length is bounded by lung capacity. Far from being a handicap, this physical constraint almost always produces readable phrasing, because it mechanically rules out the uninterrupted streams lasting dozens of seconds that only an instrument can produce. The pedagogy of singing and breath support, in fact, often defines a musical phrase, in very concrete terms, as the unit you could sing in a single breath: a definition that depends on no harmonic theory at all, only on physiology.

An instrumentalist, by contrast, doesn't have that constraint, or only partly. A pianist or a guitarist needs to breathe for none of their notes, and even a wind player, thanks to techniques like circular breathing, can in principle play without an audible break for several minutes straight. It's precisely this freedom that makes the vocal reflex so useful to impose voluntarily: nothing, on a piano or a guitar, stops you from playing an uninterrupted torrent of notes from the start to the end of a chorus, which doesn't mean that torrent produces a discourse audible to whoever's listening. Without the physical constraint that forces a singer or a wind player to stop, the instrumentalist has to impose it on themselves, consciously, or else their solo loses the readability that regular breathing gives the human voice for free.

Phrasing in breathable units, then, doesn't mean a pianist has to physically catch their breath, but that they should treat each phrase as if it had to fit within a single breath, with the same length limit and the same need to stop before moving on to the next one. In practice, this comes down to imposing a pause on yourself, even a brief one, after every group of notes that forms a complete idea, rather than letting phrases run into each other without ever marking where one ends and the next begins. A simple exercise for recovering this reflex, for an instrumentalist who's lost the habit from playing an instrument that doesn't require it, is to sing your own solo, or failing that, imagine it sung, before replaying it on your instrument: any phrase too long to sing without catching your breath is, almost always, also too long to stay readable to a listener's ear, whatever instrument is playing it.

This reflex connects directly, from a different angle, to everything this guide has detailed so far: the motif fits within a breath, the call and the response are often spread across two separate breaths, and the silence between two phrases is, most of the time, nothing other than the time it takes to inhale, real for a singer, purely mental for an instrumentalist who chooses to impose the same limit on themselves.

Building a chorus that climbs: form at the scale of the whole solo

Everything covered so far, motif, call and response, silence, guide tones, phrasing tension, phrase beginnings and endings, applies at the scale of a phrase or a few bars. An entire chorus, whether it runs twelve bars over a blues form or thirty-two bars over a jazz standard, is also built as a form in its own right, with its own trajectory from the first beat to the last, and not merely as a succession of phrases independent of one another.

By far the most commonly used trajectory is an ascending one: a chorus generally starts in a moderate register, with more spaced-out phrases and relatively low note density, then progressively builds toward a peak, most often reached in the final third of the chorus, where the register climbs, the notes tighten rhythmically, the silence between phrases shrinks, and the overall intensity, whether measured in volume or simply in the density of attacks, clearly exceeds that of the opening. This construction is nothing arbitrary: it exploits the same logic as the build of a song's final refrain, detailed in this collection's guide on song structure, except that here it applies inside a single solo rather than across a succession of several refrains. In both cases, the principle stays the same: keep a reserve of intensity for the end rather than spending it all in the opening bars.

This buildup isn't just a matter of playing faster or louder as the chorus goes on, a shortcut that quickly produces a mechanical, predictable effect rather than a real trajectory. Several parameters can climb together or separately, and it's often their relative balance, more than their direction alone, that distinguishes a chorus that's really building something from one that's simply speeding up. Raising the register without touching rhythmic density produces a different effect than the reverse, thickening the rhythm without changing register; both are legitimate, and the choice mostly depends on what the previous section of the tune just did, so as not to repeat exactly the same gesture twice in a row.

The chorus's opening motif, the one laid down in the very first bars, deserves a special place in this trajectory. Bringing it back, transformed by one of the development operations detailed above, in the chorus's final bars, closes the form in on itself and gives the listener the feeling that the solo has answered its own opening question, rather than simply stopping after a number of bars counted out in advance. This gesture, a transformed return of the opening material just before the conclusion, remains one of the most reliable ways to make a chorus feel like it has a form, a trajectory thought through from start to finish, and not just a string of good ideas with no connection between them.

What changes when you sit down to improvise

This collection's guides devoted to the pentatonic scales, the dorian and mixolydian modes, the four modes that complete them, and the ii-V-I cadence each answer, in their own way, the question of material: which notes an improvisation is allowed to draw on, at any given moment, over this or that chord. This guide has added nothing to that list of notes, and that was never its goal: it took that material as a given in order to focus exclusively on what you do with it once it's chosen.

The motif and its development give the solo a syntax, a way of stringing ideas together that goes beyond a mere succession of correct notes. Call and response organizes that syntax into dialogue, with yourself or with the other musicians in the group, rather than into a continuous monologue. Silence, far from being a lack, carries part of the meaning on equal footing with the notes surrounding it. Targeting a chord's third and seventh on strong beats turns a correct scale into a line whose harmonic movement the listener clearly perceives. Phrasing tension, independent of the harmonic tension already covered elsewhere in this collection, is built through rhythm and placement rather than through note choice. Knowing where to start and end a phrase, and consciously respecting or breaking the two- or four-bar grouping inherited from a song's own structure, distinguishes a phrase that's been thought through from one that simply stops when the musician runs out of breath or ideas. And a chorus's overall trajectory, its climb from the first beat to the last, turns a solo into a form rather than a pile-up of good phrases with no connection between them.

None of these seven gestures demands a single extra note beyond what this hub's guides on scales and modes already make available. That's precisely what makes them accessible as soon as you've chosen your scale, without waiting to have exhausted everything harmonic theory still has left to teach: the difference between a solo that recites a scale and a solo that tells a story never comes down to the number of available notes, but to what this very page has just detailed, the way those notes are organized in time.

Common pitfalls to avoid

Explore in the Music Hub

More guides