Introduction

Melody is a universal feature in most kinds of music. A simple definition of melody  might be: “a succession of notes that we perceive as a coherent whole.”  The way in which a melody is organised and how it sounds is highly dependent on its context—its historical time, place, style, and so on. In this chapter we consider mainly the principles of organisation in tonal melody in Western common-practice music, although some of the ideas are applicable outside of that (and, indeed, our first examples are not taken from tonal melodies, but come from a much earlier period of Western music).

In this introduction to melody, the main points that we cover include:

• Melodic motion
• Melodic shape
• Rhythm and metre in melody
• Tonal and harmonic coherence in melody
• Melodic motives
• Melodic repetition and and sequence

Melodic Motion

Melodic motion refers to movement (motion) from note to note. In the dimension of pitch, there are two basic principles to consider: direction of motion and distance.

• Direction of motion can be ascending or descending from one note to the next, or it may be static—that is, a note is repeated.
• Distance is measured by the interval from one note to the next.

In a melody using notes from a diatonic scale, any motion between adjacent degrees on the scale is known as conjunct motion.  Such motion will, therefore, be by tone or semitone. We also refer to such motion as proceeding by step, or stepwise motion.

Any motion between two notes that are not adjacent on the scale is known as disjunct motion.  It is also useful, especially in tonal music based on triads, to distinguish between (a) disjunct motion of a third and (b) that of larger intervals (a fourth or more).

Motion by third can be described as proceeding by skip—because we are skipping one note of the scale—e.g. C–(D)–E or B–(A)–G.

Motion by a fourth or more can be described as proceeding by leap.

Example 00 shows instances of all these kinds of melodic motion. It is a transcription into modern staff notation of a medieval antiphon by Hildegard of Bingen (c. 1098–1179), one of the most prolific composers of this kind of music in the High Middle Ages.

A brief notE on Medieval Chant notation

The original Medieval chant notation in which the music from Example 00 was written down used a system of signs called neumes. These signs evolved over centuries, and at first indicated pitch direction, initially without a staff, serving more as an aide memoir. Later, this notation system acquired early forms of the modern staff. The music from which Example 00 is transcribed is shown in its medieval chant notation in the image below.

The neumes are the various markings in a variety of shapes above the text, which is set underneath a four line staff.  The red line, which is the second line of the staff, denotes F, a fifth below C, which is indicated by the small C (an ancestor of the modern C clef) at the start of the top line. The content of Example 00 takes up about the first quarter or so of the line of chant shown shown above.

This style notation conveyed less (as far as we know) about rhythm and note duration, at least in precise terms. Modern transcriptions, therefore, tend to use filled note heads without stems, implying little or no specific rhythmic information. This allows a unique focus on the pitch dimension of melody.

Melodic shape

While melodic motion refers to the progress from note to note, melodic shape is a more “architectural” concept of melody as a whole. Some of the factors to consider in thinking about melodic shape include:

• Structural pitches—notes which act as anchors or scaffolding for the melody
• Range—the interval between the lowest and highest notes
• Tessitura—where the melody mostly “sits” within its range
• High (climax) and low points—where the melody meets its extremities of range
• Contour—the overall “topography” or “envelope” of the melody
• Distribution of conjunct and disjunct motion

Example 00, from another piece of music by Hildegard of Bingen, shows most of these concepts quite clearly. As in example 00, this is a short excerpt from a much larger piece, and is transcribed from Medieval chant notation.

The range of this melody is a perfect eleventh, spanning A3 to D5. The highest note, or climax, (D5) is reached once, in the middle of the entire melody.  Prior to this, roughly half way between the start and the middle (or a quarter of the way through), the melody reaches its lowest point (A3). Two features tend to make these extreme points particularly noticeable: (1) They each occur only once and (2) the tessitura is much narrower, spanning a only major sixth from C4 to A4—notes within this span account for approximately 85% of the total pitch content of the melody.

The two blue lines at the bottom of the example trace the contour of the melody. The first follows the line of pitch fairly closely while the lower line is a more generalised curve. If we were to take only the starting, highest and closing pitches (D4–D5–D4) a further abstraction can be imagined which has the melody as a simple arch spanning an octave (D4–D5–D4) as shown by the dotted curve, but the ascent to the high point is more complex than the descent. These are all careful choices made by the composer.

Structurally, we can easily conclude that the pitch classes D and A play a powerful role in providing scaffolding for the melody. D4 is the starting and ending point, and D is the final of the mode (D Dorian, with a brief Aeolian inflection), and A4 is the focus of the initial leap and then a point of lingering in the long descent from the climax. (A is also the dominant of the mode.)

Another distinctive feature is the way in which leaps are “filled in” by returning in step- and/or skip-wise motion in the opposite direction to the leap. This seen both in the very first part of the melody and also in the sense that the entire second half of the melody is a long and gradual filling in of the bisected octave leap D4–(A4)–D3 which reaches the climax point.  The upward leaps generate a tension which is slowly relaxed by the following slow descents back “inside” the leap, coming to rest each time on D4 (the modal final).  Also, the initial leap of a fifth lays the ground for the much more dramatic leap up by an octave (via the fifth).

The melodic motion in Example 00 is overwhelmingly stepwise, accounting for more than half of all note-to-note progressions in the melody.  This is quite typical of this style of music, but it is also the case that most melodic styles prioritise step-wise motion and use leaps to create a kind of scaffold which can be elaborated and filled in. All four types of motion and their frequency of use is shown in the table below.

Finally, because this music has a text, we should briefly consider the words: “O antiqui sancti, / quid admirramini in nobis?” (Oh ancient holy ones, / what do you  admire in us?) The melody is in two parts (indicated by the small mark on the top staff line after “sancti”). These two parts clearly correspond to the two halves of the line of text.  The first half is an address to the “ancient holy ones” with the exclamation “Oh” receiving the initial leaping gesture.  The second half is the question (“what do you admire in us?”) with the climax falling on the accented syllable of the word admirramini (admire).

More Detail: leaping, bisecting and filling in

Before adding the considerations of rhythm and tonality to our exploration of melody, it is worth setting out some general principles of melodic writing in relation to leaps. Please note, however, that what is “right” or “wrong” in melodic writing, or what makes a melody “good” or otherwise, is highly style specific. There will be plenty of occasions in which these principles can and should be set aside.

Principle no. 1: Leaps should be made by consonant intervals—avoid sevenths, ninths, diminished and augmented intervals (of course, there will be valid exceptions, often for expressive purposes—the opening tritone in Bernstein’s “Maria” from Westside Story being a famous case in point.)

Principle no. 2: It is a good idea to “fill in” a leap by following it with stepwise motion in the opposite direction to the leap. (We sometimes describe this as coming back “inside” the leap.) This can be a complete fill, returning to the note that was leapt from, or a partial fill, returning some of the way.

Principle no. 3: Larger leaps (fifth and higher) can be “bisected” by adding intervening notes.  As a rule, it is best to “divide” the leap as evenly as possible. For example, make a fifth into two thirds, rather than a second and a fourth (or vice versa). (See, for example, the octave D4–D3 in Example 00, above.)

Principle no. 4: You can fill in one of the bisecting intervals with steps, but it is mostly preferable to fill in the upper interval rather than the lower one. (This can sometimes be ignored in descending contexts—a famous example being the opening of Tchaikovsky’s First Piano Concerto.)

Example 00 shows some of these principles in a non-rhythmical context. Metric placement and rhythmic values will obviously affect the choices here. As a general tendency, the note that is the goal of the leap takes a relatively strong stress, the note of origin might be either weaker in stress or metric placement, or equal.  Again, this is not a “rule.”

Melody with Rhythm and metre

The examples we have so far considered do not involve explicit rhythmic or metrical elements. That is not because they have no rhythm, but only because many of the specifics of rhythm are either not captured in the original notation of these melodies or we no longer know what indications within the notation refer to rhythm in more than the most general terms.

The development of the metrical and rhythmic character of Western music evolved over several centuries beginning with the first emergence of mensural notation in the late thirteenth century. By around the beginning of or just prior to the tonal common-practice era (that is, in the early seventh century), as tonality was starting to coalesce into a new system that would replace the old modal system, musical time was also evolving from the mensural system of the Renaissance period into the metric system with its (more-or-less) familiar time signatures and the more frequent and regular use of bars and bar lines.

Rhythm and metre interact with pitch in melody in a variety of ways. Example 00 shows three versions of the same melodic shape. The middle line (b) shows the original form of this melody as composed by Clara Schumann. The upper line (a) is a “de-rhythmicised” version, showing pitch alone. The lower line (c) is our recasting of Schumann’s melody with different rhythms in a different time (compound duple compared to Schumann’s common time).

An interesting aspect to this comparison is that the “de-rhythmed” melody in (a) has a kind of latent quality that is waiting to be brought into sharper relief. In the two versions with rhythm, the quality between them is different, but not so much that we hear them as unrelated. If you take the opposite approach, however, and remove the pitch content and compare the rhythms of each melody, it is not really possible to hear any connection (aside from having the same number of “events”).

Rhythm plays an important role in creating what we might call the “character” of a melody, in drawing our attention to certain features, or showing those features in a different way. Consider the material under the bracket marked x in versions (b) and (c). In both there is some sense of mounting tension as the. pitch contour rises steadily from G to B. But this has a more determined quality in Schumann’s version (b), helped by the upbeat D pushing on to the strong beat on G, then through A in the middle of the bar to B on the first beat of the next bar.  By comparison, in our version (c), this segment has a more breathless quality, with its interposed rests and the fact that different notes fall on the strong beats.

The simple duple (or quadruple) time in (b) gives it an a steadiness more akin to walking, while the compound duple time of (c), with its triple subdivision, lends this version a more dance-like, or skipping feeling. This is enhanced by the fact that (c) starts with an anacrusis while (b) commences on the down beat. This difference also makes D the focal point in the first bar of (b), whereas this role is more likely played by G in (c).

It is not hard to hear in (a) that G and D play the main structural roles, not least as forming the tonic and dominant of the G-minor scale. However, in the two rhythmic versions, there are some important differences in the ways these structural pitches are emphasised. Both melodies are easily divided into two parts. In (b) the first part ends with the A in bar 4 and the second part commences with the D that forms the anacrusis to the G on the first beat bar 5. However, this same D seems to be the final note of the first part of the version of the melody in (c). In (b) the final structural pitch—the melodic goal, if you like—is D; in (c) it is the penultimate note, G, with the D more like an afterthought.

We will talk about harmonic implications further down in more detail, but these shifts in rhythmic focus and stress have implications for how we hear the melody harmonically. In (b), Schumann’s original version, the final harmonic support for the D is the clearly the dominant triad (V). In our version, the last two bars with G and D would likely be harmonically supported by the tonic triad (i). (The harmonic analysis in Roman numerals below (b) and (c) is there to show the basic and likely indications only.)

Agogics

Metre and rhythm can combine in quite powerful ways to place structural emphasis on certain notes of a melody in contrast to others. The example below, which comes from the early seventh-century, is a simple case in point. This short passage comes from the opening instrumental ritornello from an aria for solo voice with basso continuo accompaniment; only the melody is shown here. The minimal metrical structure is shown by the dots below the staff, where the tactus is identified as a pulse in minim values with a duple subdivision.

The melody consists of a series of short descending shapes, moving almost always in conjunct motion. These descents are shown by the sloping arrows. After each of these shapes, there is a leap or skip up to a quaver upbeat which initiates the next shape (which fills in the leap or skip). Notice how each shape ends with either one or two longer values, coinciding with the tactus pulse. (We can ignore for the moment the very brief upbeat quavers that take some value away from previous long value to form dotted crotchets.) These longer values add weight to the end of each shape by placing an agogic accent on these notes. An agogic accent is where a musical note is emphasised by its comparative length, as compared to its loudness.

A distinctive feature of these agogically accented notes is the way in which they relate to the scale of G minor, on which the melody is based. In bars 1–4, these notes are found to be the leading-note and the supertonic, which both point logically by step to the tonic. In bars 5–8, the agocially accented notes form a descending scale segment from the dominant pitch () down by step to the tonic .  This is all summarised in the melodic abstraction at the top of the example, in un-measured note heads. This highlights a simple architecture for the melody based on the agogically accented notes, working to prioritise the tonic degree of the scale.

Like a lot of early Baroque music, the rhythmic organisation in this melody takes something from classical poetic metre and the associated concert of a metrical foot. A metrical foot is a classification of groups of syllables according to length (in classical Greek or Latin) or stress (in modern languages, such as English). Two symbols are used to indicate these qualities: the macron (–) for a long or stressed syllable and the breve (◡) for a short or unstressed syllable.  In various combinations, these form the various “feet” of poetic metre, such as the iamb (◡ –) or the dactyl (– ◡ ◡).

In the melody in Example 00 the arrangement of short melodic shapes suggests two types of foot: anapaest (◡ ◡ –) and double iamb ( ◡ ◡ – –).

Rhythm can sometimes introduce agogic accents in a melody that are in conflict with the metre. This is not necessarily an error, but rather a use of rhythm to make the melody somehow distinctive and memorable. When introduced, such devices are best used consitently rather than at random. In the following example, in duple metre, a set of longer note values (minims) land on a subdivision of the tactus rather than the tactus beat itself. In one case, the B in the third bar, the long duration falls on the next smaller subdivision (quaver).  In the dots below the staff which indicate different pulse layers in the metre, the beats where no melodic note falls are shown in a lighter shade. The character of the melody is that of a brisk march and these off-set agogic accents generate an added sense of vigour. In a different way to Example 00, the agogically accented notes also play are role in emphasising tonality—they are all members of the tonic triad, E major.

Rhythmic grouping

Rhythm and metre also provide indications about smaller groupings of related pitches in a melody. In Example 00 below we have sixteen bars of melody. It is “in” simple triple time, although a more complete representation of the metre is shown in the ski-hill graph underneath.

If we take a “top down” view, the melody can be understood to divide into two units of eight bars (phrases). A rhythmic feature of each phrase is that it begins with a crotchet anacrusis which establishes a pattern of linking the final crotchet of a bar to the down beat of the next. Further down, each phrase divides neatly into two groups of four bars, and these are marked with horizontal brackets denoted a, b or b’ (which is pronounced “b prime”). Note that in both phrases a is identical while b and b’ are similar but not identical, hence the differentiation in labels. We can also imagine a fairly easy further division of a into two bar units (marked x and y). However, this is much harder to do in b or b’.

The markings below the line reflect a division according to stressed and unstressed beats relative to the metre, at the level of the tactus. Note here that this is not accounting for agogic stress, but metrical position. Therefore, according to the metre and the anacrusis, this sets up a recurring pattern: ◡ | –  ◡ (an “amphibrach”).

In the first half of each phrase (a), the melody has a clear and strong shape. Overall it forms an arch from A4 up to A5 and back to A4, the dominant degree (). The first two bars (x) involve an in-filled leap of a sixth, and come to end on the supertonic (). These bars use exclusively crotchet values and establish the tactus and the metric “foot” that straddles the bar line, consolidating the  ◡ | –  ◡ stress pattern.

The second half of a (y) introduces quaver motion, and the effect here is to create something of a smoother link between the two metric feet. Nonetheless, the overall rhythmic and metric order remains well established.

The situation in b and b’ is quite different.  Firstly, the pitch profile of the melody in these bars is much narrower compared to the melody under a.  The range is reduced to a fourth (in b) and a third (in b’). The line generally meanders a step either way around E5 (). Moreover, the rhythmic grouping changes such that the unstressed (◡) value  at the end of bar 5 feels connected to the quaver group that precedes it rather than as an anacrusis to the next bar. This effect is created by shifting the rhythmic group of four quavers from beats 2 and 3 in bar 3 to beats 1 and 2 in bar 5 and reinforced through repetition in bars 6 and 7.

Therefore, as the melodic shape “flattens” our attention is drawn to the subtle rhythmic disruption and the implicit change not in musical metre but in the grouping of stressed and unstressed beats over the metre. The bracketing under the poetic metre signs (◡ and –) represents our best attempt at showing this grouping. Neither b nor b’ are identical in this distribution, but there seems to be an organisation into three units of unequal length in both phrases. Note, again, that the musical metre does not change, but the arrangement of stressed and unstressed notes and their groupings is affected by simple rhythmic changes.

Tonal and harmonic Properties of Melody

In tonal common-practice music, melodies can be expected to express a relationship to the key and to be organised around harmonic frameworks, such as triads, and harmonic progressions. Although this chapter precedes discussions about functional harmony, we can mention a couple of simple ideas on the topic that will be enough to understand the points made below.

Phrases and basic harmonic structure

Depending on the style, a melody will often divide into related components that have a degree of structural independence.  There are many ways this can happen, but one of the simplest and most pervasive in Western common-practice is found in the various kinds of archetypes used by Classical and early Romantic composers, especially in smaller compositions or as the starting point for larger ones. These are known as themes.^[1]

A theme in Western common-practice music is a clearly identifiable set of melodic ideas, and is always bounded by at least one clear point of termination, called a cadence. The most standard types of theme are typically eight bars long, and usually divide into two four-bar subsections called phrases. Phrases also often terminate with an identifiable cadence, although in formal theory (as compared to harmonic theory), this doesn’t always have to be the case. We will look at two examples, to show how this relates to melody.

In Example 00, a fairly simple melody in F major is structured around the tonic and dominant harmonies, which act as starting and ending points. The scale-degree numerals above the staff indicate the main structural pitches of the melody and the Roman numerals below indicate the main chords implied (all either I or V).

These eight bars form a theme which readily divides into two groups of four, which can be identified as phrases.  Both phrases begin the same way—a compositional strategy we encounter frequently. In this case they each share a short, two-bar figure we call a basic idea (b.i.).  This is complemented in each phrase by a further two-bar figure which we call a contrasting idea (c.i.). Looking at the first four-bar phrase, both the basic idea and the contrasting idea are structured around a melodic frame of leading to above an implied harmonic progression from I to V.  Listening to this, an expectation is now established that ultimately the melody is destined to finish on , but so far is not able to reach that goal. The arrival on in the fourth bar provides a point of partial repose, but not complete closure.

In the second four-bar phrase, the basic idea is the same, but the contrasting idea is modified, by compressing the harmonic rhythm and having a structural progression in the melody of –– over an implied I–V–I harmonic progression. At this point, the sense of closure is fulfilled. Both phrases conclude with what we call a cadence, but the second cadence provides a greater degree of finality than the first. We often use the terms half cadence (HC) to describe the first and perfect authentic cadence (PAC) to describe the second. The full meaning of these terms is explored in the context of harmonic cadences, but harmonic cadences often support melodic shapes such as we see at the ends of these phrases.

The two four-bar phrases in Example 00 comprise a particular type of eight-bar theme called a period, in which the first phrase, called the antecedent, expresses a sense of partial melodic and harmonic completion (marked by the half cadence) and the second phrase, called the consequent, expresses a sense of full melodic and harmonic closure (marked by the perfect authentic cadence).

A different kind of theme is shown in Example 00. It is similar to Example 00 only in being eight bars long and having two four-bar phrases. The nature and function of the two phrases, however, is quite different here. Together they comprise a type of theme called a sentence.

The first phrase here is made up of a basic idea which is immediately repeated, in a varied form. In this case, the basic idea is extremely simple, and the two iterations create a slowly rising contour from to and then to . Notice that the first basic idea’s — is supported by the harmonic progression i–V and the second basic idea’s – is supported by the reverse, V–i.  The effect is a little like a question and answer, or, perhaps a comment and a response.  Despite this effect, there is no strong sense of cadence. Altogether, these elements make up what is called a presentation phrase.

In answer to this, the next phrase offers a more continuous four-bar group, much less readily divisible into two. This is known as the continuation phrase, and this kind of phrase always leads to a cadence. In this case, we end with a half cadence with the final main melodic note, the leading note G, implying the dominant triad (V). Note also how the continuation phrase, as if in response to the presentation phrase’s rising contour, outlines a scaler descent from to .

While we have not shown the other parts (the accompaniment to Example 00 and the left-hand part to Example 00), it is quite clear that these melodies work hand-in-hand with simple harmonic supports.

Harmony and non harmony notes

Some melodies can be built entirely from just the notes of chords—we call these harmony notes. The following is a very well known case of a melody built entirely from harmony notes—Example 00.  The key is B major and almost every degree of the scale is used, except for the submediant (G).  The melodic motion is comprised mostly of skips and leaps, moving through the notes of either the tonic triad (B–D–F) or the dominant seventh chord (F–A–C–E). The only places where the melody moves by step is as it goes from one chord to the next.

Of course this melody comes from a song with piano accompaniment but even without this accompaniment the sense of the harmonic progression involving only I and V⁷ (or just V) is very clear. This progression is shown by the Roman numerals beneath the line. We do not need to hear every note of a triad in the melody to understand what triad or seventh chord is implied.  While bar 6 contains every note of the tonic triad, in bar 5, the first beat only includes the fifth and root of I, and the second beat only gives us the third and seventh of V⁷. None-the-less, the implication of these chords is quite clear.

Also important to note is the rate at which the implied harmony changes between the I and V⁽⁷⁾ chords—this known as harmonic rhythm. Underneath this melody, it moves steadily and rather slowly in crotchets and minims, while the actual rhythm of the melody—what is often called the surface rhythm—moves faster, in shorter values, and is more varied.

Melodies such as Example 00 are comparatively rare. Most melodies contain at least some notes that sit outside the what is the active harmony at any point. We call these non-harmony notes. In the simplest of scenarios, we can find very clear connections between harmony and non-harmony notes. Example 00 continues directly from Example 00. We first pay attention to bars 13–16 (the first line).

In comparison to most of the melodic motion in Example 00, the melodic motion in bars 13–16 is entirely conjunct. This does not mean that the chords change with every note—far from it.  In fact, the harmonic rhythm has slowed down such that it progresses in two-bar units, moving from vi (G–B–D) to V (F–A–C). The basic descending triadic shape of the melody is shown on the smaller staff above.

The conjunct motion in bars 13–16 arises from two categories of non-harmony notes:

• Passing notes, which fill in skips (thirds), to create conjunct motion between harmony notes. (Occasionally, although not in this example, a pair of passing notes can fill in a leap of a fourth.)
• Neighbour notes, which embellish a single harmony note by conjunct movement away from the note and back to it.

These are annotated with the abbreviations P and N in Example 00.

The basic shapes of melodic motion using passing and neighbour notes is always conjunct. In Example 00, these basic shapes are summarised—harmony notes have stems, non-harmony notes (P or N) are given in note heads alone. Like the skips or small leaps they fill in, passing note patterns can be categorised as ascending and descending.  Neighbour notes that move up by step and back down to the harmony note are called upper neighbours, the opposite type are called lower neighbours.

To conclude the discussion of Example 00, we also find mainly conjunct motion in bars 17–20.  However, most of these notes are heard as harmony notes (apart from the upper neighbour E).  This is because the original harmonic rhythm of predominantly crotchet motion returns, coinciding with the predominantly longer rhythmic values in the melody.

The next example includes some slightly more complex non-harmony notes. The melodic surface is varied but note how the underlying harmonic rhythm is quite slow, each chord being active for either a whole bar or for two bars. (The analysis here is somewhat simplified and does some details that are in the accompaniment part, which is not shown.)

Bars 5–6 are fairly clear, with use of passing and neighbour notes.  In bar 8, the C (marked ?) is hard to categorise with the definitions we have so far. provided  It partially fills the gap between A and D, but it is not really a passing note because the motion is not fully conjunct—A to C is a skip.  We will find many more instances of this kind of thing, and in this case we can categorise it as an incomplete neighbour note.  Because it moves by step to D, it is understood to be embellishing D.  Also, because the alteration from C (which is diatonic) to C (which is chromatic) reduces the interval between these notes to a semitone, the connection is made even stronger.  If you listen carefully to this, you will notice how the C “leans” into the D, acting like a “mini leading note.”

Bars 9–10 are built around the dominant seventh chord (V⁷). In fact, three notes of V⁷ are present—the third (B), fifth (D) and seventh (F). Only the root (G) is not present in the melody (it does appear in the accompaniment, which we can’t see). Moreover, looking at the shape of this part of the melody, it is bounded by F and D and we can think of the E at the start of bar 10 as a passing note between them.  However, before this E can appear, the melody leaps down a diminished fifth, to B (the third overall V⁷). This leap is then filled in by motion to C, which can now be thought of as a passing note from B to the final note, D. Further complicating things is that this C passing note appears to have its own neighbour note (B).  Of course, B is already part of V7, but in this very local context, it can sound like a neighbour to C.

These two bars are shown in an analytical way in (a) and (b) at the bottom of the example. In (a) all the steps shown above are illustrated using stemmed note heads for the harmony notes and non-stemmed note heads for non harmony notes.  Arrows try to account for motion. In (b) a much more abstract account is made, in which we can understand these two bars as two melodic progressions moving via passing notes in contrary motion.^[2]

Motives

Motives are small groups of notes that form an identifiable unit within a larger melody and which can be used as “building blocks” for the melody.  Motives are generally short, a few notes, and are used in different ways to lend a melody a sense of cohesion.  Motives typically have both a distinctive melodic and rhythmic content.

The melody in Example 00 has two motives introduced in the first two bars—the rising sixth, and then the descending group of notes under the bracket marked a. Of the two motives, the second one turns out to be the more extensively used—see the brackets marked b to f, all of which are related to the first appearance of this motive under a. We might say that they are variants or derivatives of the original form of the motive.

The variant of the motive at b is simply a diatonic transposition a step higher of the motive at a. At c we find that the initial descent of a, E to C, is “levelled”, simply repeating the note C, retaining the rhythm only . At d we find the direction of motion is changed from an overall descending line to an ascending one. Effectively, the motive is turned “upside down”—we call this technique melodic inversion. (This is a different, although related, practice to inverting intervals or triads). Additionally, one of the notes is “chromaticised”—D becomes D. At e, we find a transposition of the variant first introduced at c. Finally, at f, we find the most altered form of the motive.  It is best considered a further variant of c. The repeated note, now being D, is elaborated by lower and upper neighbours and the final two notes proceed by a rising leap rather than descending step. Perhaps some might find this final version as being too remote to be accepted as a variant of the original. On the other hand, we might argue that the gradual process of changing the motive has led, by a logical sequence of steps, to this final version.

The effect of constructing the melody in this way is to give it both cohesion and variety. The term used to describe this process of changing a motive while retaining its basic identity is development.

Melodic repetition and sequences

While repetition might suggest a lack of creativity, melodies often use repetition in order to lend a sense of coherence and belonging. You can probably already identify some instances of repeating material in some of the examples above. Repetition can involve motives or larger units of melody.  It can include very minimal variation to a melodic unit, without substantially altering identity. Finally, a particular type of repetition can involve transposition of a melodic unit—this is called melodic sequence. Example 00 shows examples of all these kinds of repetition.

In these sixteen bars, the first eight involve four statements of a basic two-bar figure, labelled x. This is repeated immediately with a very slight variation, x’. The four-bar combination of x + x’ (labelled X) is then itself repeated. A further variant of x, marked x”, commences with the anacrusis to bar 9, an octave higher.  This is then repeated a third lower, meaning bars 9–13 form a sequence. We term the first part of a sequence, which states the figure to be transposed, as the model.  In this case the sequence consists of only one transposition of the model, but sequences can have more than one, provided that the interval of transposition remains constant.

In this example, it seems that bar 13 will initiate a further transposition a third lower again, but the process is altered. Only the first five notes are used and a shorter version of x”, which is labelled x”’, initiates a new model which is sequenced once, this time a second lower.  Bars 9-14, therefore, contain two melodic sequences.  Each has one transposition of its model—the first down by a third, the next down by a second.

The transposition of x”’ in the second sequence has a short extension, but this does not invalidate the sequential relationship.

Finally, the melody ends with a cadence figure, labelled y. This figure is repeated immediately an octave higher, creating a kind of echo effect, lending emphasis to this new idea and reinforcing the sense of partial melodic closure. A repetition that is transposed by an octave (or multiples of an octave) does not create a sequence. Such repetitions do not involve a change of pitch class.

The shape of this melody is quite distinctive. Considering the first part (bars 1–8), you might wonder at first if it has sufficient variety, if the repetition is excessive and lacks interest. It seems to be “stuck” on the iterations of x, meandering around and lacking direction. However, seen in relation to the whole, there is a justification—the obsessive nature of these bars intensifies, through contrast, the effect of the sudden launch up with the anacrusis to bar 8, initiating a long drift downwards (utilising the sequence technique).

Of course, another factor to consider is that we have looked at the melody alone. So far, we have refrained from supplying much harmonic context to the melodies we have examined (save for some basic harmonic implications). However, it is worth considering this melody in the full context. While a harmonic  analysis is provided, do not pay much, if any, attention to the details. Focus instead on the sound.

Two things are clear from this, in general. In bars 1–8, where x is stated (with variation) four times, each statement is reharmonised. So, the variety the melodic repetition comes in the form of the underlying harmonic support. By contrast, in bars 9–14, with the two melodic sequences, you will notice that the harmonisation is locked to the sequence. By this we mean the profession under the model is replicated under the transposition at the same interval of transposition. For example, the iv on the the first beat of bar 9 becomes ii^ø on the first beat of bar 11. The model is transposed down a third, ii^ø is a third lower than iv. You can follow the same “mathematical” relationship for each analogous chord in the model and the transposition, changing the interval to a second for the shorter sequence in bars 13–14.