A beautiful sound is the basic raw material of music. With this raw material, we produce notes – the building blocks (bricks) of music. But once we know how to find our notes and make them sound beautiful, we then need to organise them, structure them, assemble them together into meaningful groups which we call phrases. Just as with spoken languages, the word “phrasing” refers to the the way in which we connect sounds to each other in order to make, first of all, smaller groups (the equivalent of words), and then larger groups (the equivalent of sentences). This is one of the ways we give sense, structure, direction and intention – and thus meaning –  to what would otherwise be just a succession of (hopefully gorgeous) noises. It is largely through phrasing that we transform a succession of notes into a work of art that is not only beautiful but also interesting, dramatic, powerful etc. Phrasing makes music (the notes) come alive.

In the hierachy of challenges that a performing musician must overcome, phrasing – finding “meaning” in the music – represents the ultimate test of musicianship. Making a beautiful sound is perhaps the first, most basic challenge, while smoothly connecting the many different beautiful sounds (notes) one after another (technique) is the next level. This triple-hierarchy has a definite parallel in the worlds of psychology, philosophy and religion. One of the most primitive human drives is towards the obtainment of physical pleasures (Freud).  At the instrument, this search for physical sensual pleasure corresponds to our search for a beautiful sound. On a slightly higher level is our drive towards power, money, status (Adler). This level corresponds to our technical prowess. Our ultimate human happiness however comes only when we find “meaning” (Victor Frankl). Meaning in music comes from phrasing.


There are two levels of contribution to the phrasing of a piece of music. Firstly, is the composer’s contribution, and secondly, the performer’s. We, the players, are not making the phrases from zero: we are not “inventing” the phrasing.  The music has its own implicit, inherent phrasing and usually composers write a lot of instructions such as dynamics, articulations, slurs etc to help us bring out their phrasing concepts. But just following these instructions is not usually enough to make a really great interpretation – there is always room for more ideas. This is especially the case in older music because, normally, the older the music, the less instructions we are given by the composer. The written music that composers have left us is like an actors script that we must bring to life by applying all our interpretative skills and judgement in order to extract from it (or simply uncover) the most possible meaning.

Whereas producing beautiful sounds is very much a physical job, organising these sounds into phrases is a considerably more intellectual activity. Some players do it brilliantly with intuition (Jaqueline du Pre), others need to consciously think about it more. But independently of how we get there, it is this ability to find or create direction and meaning (rather than just following instructions), that makes the the difference between a bricklayer and an architect, an assembly line worker and a designer, an artist and a technician, a creative communicative instrumentalist or a mechanical one.


Dynamics are a large part of phrasing –  but they are not the only tool we can use. For the player, as for the actor, converting black spots on a page into meaningful phrases will normally require the use of our entire interpretative tool box. We make and differentiate phrases not only with dynamics but also with our choices of articulations (including bowings), rhythmic freedom (see rubato and rhetoric) and colour (including fingerings).


As with the origins of musical articulation, the origin of phrasing is speech. In speech, the standard phrase unit is a sentence. This corresponds in music to the standard three or four bar phrase. Phrasing however is not limited exclusively to complete musical sentences. In fact, each bar can be a mini phrase (like a clause in a longer sentence), and in fact, each single note (like a single word) can be a micro phrase.

The way we treat the dynamic (volume) of each note after its beginning is part of our phrasing. Unless the note is very short, we can usually choose from the following dynamic possibilities: diminuendo, hairpin, crescendo, sostenuto. Of course, the longer the note the more we can do with it and the more important it will be to actually do something. Playing slow music makes us really think about phrasing as we can’t hide behind constant note movement to provide interest. Look at the following example from the beginning of the Bach-Gounod Ave Maria. Here there are only 5 notes in the 4 bar phrase and each note (bar) can be phrased differently to good musical effect.

bach gounod new


We tend to remember our telephone number with the digits grouped always in the same way. If however somebody says our telephone number to us with the digits grouped differently to the way in which we remember it, it usually sounds like a different number and we are unable to recognise it as our ultra-familiar number, for example:

499 48 70      as     49 94 870

65 28 54 17     as      652 85 417

   427 518      as        42 75 18

The same happens with musical phrasing – if we play with the “wrong” phrasing, we can convert a masterpiece into nonsense music without actually changing any of the actual notes.


The need for all these tools is made very obvious when we look at certain pieces in which hundreds and hundreds of notes of the same rhythmic value follow each other uninterruptedly. Bach was a champion of this type of music, and the Preludes of his Solo Cello Suites provide some very good examples of this rhythmic “monotony” that so desperately needs to be structured, organised, broken up into phrases:

  • In the Prelude of Suite I, of the total of 657 notes, only 3 are not consecutive semiquavers.
  • In the Prelude of Suite III, from bars 25 – 77 (i.e. everything that comes between the introduction and coda) we have 895 uninterrupted semiquavers.
  • The Preludes of Suites IV and VI start respectively with 384 and 950 uninterrupted quavers.

Even if we play all these notes perfectly and beautifully, these wonderful stories that we are telling can sound horribly monotonous if we don’t do any phrasing. Playing beautifully but without phrasing is comparable to somebody with a wonderful rich voice, reading aloud from an exciting book, but with no interest or understanding.

The Bach Solo Cello Preludes – and in fact Bach’s compositions in general – are a wonderful test of phrasing ability for performers (and conductors). Not only do the notes often come in constant succession with little rhythmic variety, but also Bach gives almost no phrasing instructions at all apart from the occasional bowing. In fact, playing Bach is almost like an IQ test for musicians in which we are asked to “make as much sense as you can” out of long sequences of rhythmically identical sounds on a page. For more examples of this phenomenon in the music of Bach, look at the following table referring to the lack of rhythmic variety in 5 of the 8 movements of Bach’s b minor Partita for Solo Violin.





Double I


378 (semiquavers/16th notes)



474 (quavers/8th notes)

Double II


953 (semiquavers)

Double III


298 (quavers)

Double IV


531 (quavers)



Musical phrases, just like phrases in any language, come in an infinity of varieties. They can be regular, irregular, separated or overlapping, coinciding with the bar lines or displaced with respect to the bar lines. And of course a lot of modern classical music deliberately has no meaningful phrases. Phrasing reached its peak of clarity and regularity in the Classical Period with its emphasis on order, balance and clarity. This is why music of this epoch is probably a good starting point for a study of  phrasing.


We can find lots of a nice parallels between dance choreography and musical phrasing. When we are playing unaccompanied music, there is only one line of music to phrase and we don’t need to organise our phrasing with anyone else – it is as though there was only one dancer to choreograph. But when the music has more than one voice, this is like having more dancers on the stage and now the possibilities for phrasing (choreography) start to get more complicated. When, in our piece of music, all the musical voices are phrasing identically together, this corresponds to the dancers all doing the identical movements at the same time. But at other times the dancers (and musical voices) might all be totally independant from each other.  And in between these two extremes of total union at one end and and total independance at the other,  there exist a huge variety of possible combinations  (choreographys) of the multiple voices: sometimes there are just two dancers weaving in and out of each other (pas de deux), sometimes they copy each other with a time delay (canon), sometimes multiple dancers collide, combine, pile up on top of each other, run through each other like waves, and sometimes one is the dramatic expressive soloist while the others are just the quiet backdrop (accompaniment) etc. Phrasing is musical choreography and we would do well to imagine how we might choreograph any piece of music we are playing.


One of the most basic properties of any phrase is its length. By far the most common phrase in music of the Classical Period is the “four-bar phrase”. Often, when counting a long multi-bar rest in music of this period, we don’t even need to count the individual bars in the rests but rather just use multiples of four and know in which part of the phrase we enter. In fact, the 4-bar phrase is so common in this period that anything else can seem like a surprise trap, into which we will easily fall when playing “by instinct”. Look for example at the beginning of the Mozart C Major String Quintet, with its 5-bar phrases.


Music is a sonic representation of life. For a phrase –  just like for a life, a note, a piece of music, an event or a relationship – there are two critical defining points: the beginning and the end. Phrases are born and die. They can start (and/or finish) imperceptibly and with infinite gentleness – like dawn, sunrise, sunset, a caresse. At the other extreme, a phrase can start (and/or finish) with total violence and abruptness –  like a slap, a gunshot, a light switch going on or off. In between these two extremes there is the whole palette of different types of births and deaths: hesitant, laboured, determined, impatient, peaceful, agonizing  etc.


Phrasing – the rise and fall of the music – is what makes music come alive. Well-interpreted music is like a living creature, and phrasing is what gives it this “aliveness”. But phrasing in music doesn’t just correspond to life, it corresponds more precisely to the essence of life: to breathing. The magical transformation by which phrasing converts simple sounds and notes into something so much more meaningful,  is comparable to the transformation of simple “matter” that doesn’t breathe (the inanimate world) into  matter that does breathe (the living world). This comparison (analogy) between breathing and phrasing is worth exploring because they have so many similarities with each other. This subject is discussed in detail on the page dedicated to breathing.


Going one step deeper still, phrasing corresponds to one of the most fundamental, essential phenomena of not just the living world but also the physical inanimate universe: waves. A phrase corresponds almost identically to a wave. Waves are everywhere but let’s look now specifically at ocean waves as these are the most visible waves. A wave, when it reaches the shore, rises up, reaches a climax, breaks, and then dies out. This is just like a musical phrase  – or like a life. Some waves are small and playful: they rise gently and break gently. Others are big and powerful, breaking violently, explosively. Some waves leap up suddenly, but others have a long slow build-up.

Sometimes waves are disorganised, chaotic and multidirectional (as in a storm). Other times they are clean and orderly, moving forwards with a deep, slow, steady rhythm. But it is only when they reach the shore that they can reach their climax and release their energy. We can make the same distinction in music: most of the time we are in the open sea with an inconclusive rising and falling of the phrases. But then we get to the shore, where the phrase can finally reach its long awaited climax: rise up, break…….. and die. These are the climactic moments of a piece of music, and of a phrase.

It is fun to look at different music using this oceanic surf analogy. We can look at an individual phrase and ask ourselves “what kind of a wave is that?” Or we can take any individual piece (or movement) and ask  “how is the ocean behaving here?” We can even look at any composer and ask “what type of ocean is this?” Mozart’s waves are never scary. He teases us, bringing us from the open ocean towards the breaking zone (shore). The waves rise up as thought they were going to break violently on us, but then at the last instant the force dissipates and they just give us a gentle caress. His waves are almost always friendly and playful. In Beethoven’s music by contrast, the waves are almost always powerful, and the older he got, the more powerful they became. Brahms is a long, deep, rhythmic, orderly progression of big powerful waves (long phrases). Some music stays permanently in the open ocean (where waves don’t really break) on a calm day (Satie). Other music stays almost the whole time in the intense breaking zone on a rough day (Beethoven).

So it is not actually necessary to be a surfer in order to understand and enjoy waves. In fact a musician is more than a surfer …… musicians are not only riding the waves, we are making them at the same time!


All aspects of life and the universe can be (and are) represented in music. It is not surprising therefore that sometimes the music gets sick or sad, goes into outer space, experiences catatonic states, is only barely breathing, stops breathing, or even dies. In these moments, “making” a phrase (with a “natural” rise and fall) can be quite inappropriate. Often, this deliberate abscence of phrasing will be accompanied by the deliberate abscence of vibrato, both vibrato and phrasing being the two great “life-givers” in music. The beginning of the slow movement of Beethoven’s 7th Symphony is a good example of this.



When phrasing is predictable and symmetrical, it tends to be boring, or conducive to mindless trance (or repetitive dance). Military marches and disco music come to mind. Expressivity (like comedy) is very often associated with the surprise of the unexpected. With respect to musical phrasing, this means that when a particuarly expressive melodic note (or harmony) comes on what would normally be a weaker beat in the bar (or a weaker bar in the phrase) this serves to increase its expressive or dramatic effect. These melodic or harmonic points of emphasis, displaced away from the traditionally strong beats of the bar (or phrase), are called “agogic” accents.

Let’s take a very unpleasant example from the world outside of music to demonstrate the effectiveness of agogic accents. Torturers know that the effect of a punch (or any other sudden act of violence) is greatly magnified if the victim is previously lulled into a false sense of security by being spoken to in a kindly manner …… so be warned! Good composers and public speakers also use this phenomenon to good advantage.

Most commonly an agogic accent is a melodic effect. It is as though the melody, instead of being perfectly synchronised with the harmony, is now dancing around it, pulling and pushing against it in an expressive pas de deux. It is usually the harmonic movement (harmonic rhythm) that determines and maintains the regular pulse of strong and weak beats, but in agogic phrases the melodic “high points” no longer coincide with the harmonic changes. This is a potent device for  creating – and resolving – emotional tensions. With agogic accents we enter into an intimate emotional world of non-conformist (im)pulses far from the disciplined, regimented, perfectly synchronised formality of a militarised Olympic opening ceremony. With all this seething yearning and rebellion, this pushing and pulling against the status quo of the regular pulse, it is not surprising that Romantic music is absolutely full of agogic accents. In the following example, all the hairpins are Schumann’s (from the autograph manuscript):


But even music of the Classical and Baroque periods uses agogic accents also, especially when it wants to be particularly expressive.


Syncopation is also a form of agogic accent. Most jazz and pop music is just full of syncopations, which give it that relaxed soothing, flowing gentle touch. Bossa Nova is perhaps the ultimate expression of unmilitary gentleness through the use of syncopation: there is hardly a single note that starts on the beat! But even Mozart used syncopation and agogic accents quite often for heightened expressivity. The opening to his D minor Piano Concerto K466 is a good example of this.

The simple fact of emphasising the potential agogic accents in a musical line is enough to make an interpretation “romantic” rather than “classical”. To show this, let’s look now at a piece that is so much at the frontier between the Classical and Romantic styles that it can be played in both ways: Schubert’s Arpeggione sonata. By bringing out (heightening, emphasising, exaggerating) the many agogic melodic stress-points with which this sonata abounds, we can make this sonata sound absolutely romantic and just full of yearning. If, by contrast we play down the agogic accents we will instantly give our interpretation a more “classical”, restrained feel. Here is the cello’s first entrance, transposed down a fifth to make it easier to play, phrased in these two completely contrasting styles:


Even in unashamedly Romantic music, we can make it even more romantic by bringing out the potential agogic accents:

dvorak II

The above excerpts show examples of agogic (displaced) accents within each bar. Here the high points of the melody are displaced away from the first beats of the bars. Exactly the same phenomenon occurs at the level of the phrase. Here, the high points of the romantic phrase come on the second and fourth bar rather than on the traditionally-strong first and third bars. Even though the following examples are notated by Mendelssohn in 3/4 we have notated them here in 6/8, simply to show more clearly the agocic nature of the phrasing.

mendelssohn trio


Often, through the use of intelligent bowings, we can exploit the natural tendencies of the bow (diminuendo towards the tip and crescendo towards the frog) to help us with the rise and fall of the music. In all the above excerpts, we have tried to find the bowings that “go with” the different phrasings. Sometimes however, technical or musical factors will make it impossible to use the ideal bowing for the phrasing, and we have to bow “against nature”. Therefore, especially when there is no bouncing involved, it is a good idea to practice also with the “reverse” bowings (crescendo to the tip and diminuendo towards the frog) to prepare us for these inevitable situations.