Dynamics and accentuation

In this example, five possible dynamic levels are provided, but any number could be used. Alternatively, the horizontal lines dividing the tall rectangle could be omitted, and the volume level simply indicated by the height of the black portion at the bottom.

A famous example of dynamic contrast occurs in Haydn’s Symphony No. 94, nicknamed “The Surprise” from the sudden loud sound that interrupts the quiet opening melody of the second movement.

This example also introduces a refinement of our earlier method of representing repetition; see Beats. Here the opening melody is first played soft (piano) then repeated even softer (pianissimo), with a different final bar. The ordinal numbers “1st” and “2nd” refer to the first and second renditions of the melody, while a forward arrow under “2nd” indicates where, on the second time through, we skip to after the repeat bracket.

In Global Notation (as in staff notation), a dynamic marking applies to all the ensuing music until the next dynamic marking. Thus, if a passage is generally soft or moderate in volume but includes occasional loud sounds, we may have to write a lot of dynamic markings: one for each loud sound and one after each loud sound to return to the softer level. (This is similar to the issue we encountered with changes of beat division, which sometimes apply “henceforth” and sometimes to only one beat and/or layer.) To reduce the number of markings while better conveying the overall dynamic level of the passage, an alternative is to use a different kind of marking for the loud sounds: a marking that applies to one sound only and doesn’t change the overall dynamic level. This kind of marking is called an “accent.”

In Global Notation, accents are written the same way as dynamic markings, only smaller. Using accents would be an efficient way to notate, for instance, the beginning of the “Augurs of Spring” section of Stravinsky’s ballet The Rite of Spring, where a repeated chord on the strings is intermittently played louder and doubled by horns. As we are only discussing rhythm and accentuation at the moment, we’ll leave pitch unspecified in this example.

Differences of dynamics are not always as distinct as this: sometimes music transitions gradually from soft to loud or vice versa. This can be represented by using dynamic markings in which the top of the black part is sloped rather than horizontal.

The level at the left-hand edge of the black part represents the dynamic level from which the crescendo or diminuendo begins. If desired, a second such marking can be placed at the end of the crescendo or diminuendo to show the level at which it ends. For instance, our earlier example from Korean SamulNori drumming (see Pulseless onset timing), which dies away from a loud beginning more or less to silence, would appear like this.

Conversely, music that gets gradually louder can be represented with dynamic markings that slope the other way. A rather extreme example occurs at the beginning of the film version of Leonard Bernstein’s musical West Side Story. In a long crescendo like this, it can be helpful to include intermediate dynamic markings to remind the reader that the crescendo is still going on.

If desired, a more rapid crescendo or diminuendo can be specified by a dynamic marking with a steeper slope.

To specify an immediate change from one dynamic level to another, as in the classical dynamic marking fp (fortepiano), two dynamic markings can touch each other.

We also now have a way of representing the pulsation of volume in our earlier example of didgeridu playing (see Absolute pitch) - effectively a series of small accents.

Dynamic markings and accents that apply only to one layer are placed under that layer (as in our examples so far). When a system contains more than one layer, any dynamic markings and accents that apply to all layers are placed at the top of the system (thus reducing the total number of markings and amount of space needed). If these markings are contradicted by any markings placed below an individual layer in that system, the latter take precedence as regards that layer. This is similar to the way accents take precedence over general dynamic markings as regards the individual sounds to which the accents apply.

For instance, our earlier example of West African Kpegisu drumming (see Specifying other information where duration is unspecified) is played at a generally loud dynamic level, but the “press” strokes on the kidi drum are much softer. This could be specified by indicating a loud dynamic level for the whole system, but marking the dynamics of the kidi layer separately.

If greater precision is required in notating music from a recording, sound analysis software can be used to produce either a waveform graph or a line graph of time against volume - or, as the software programs call it, “intensity.” This graph can then be incorporated into the notation. But note that it will only measure the overall intensity at each moment, not the dynamics and accents in individual layers; so, as usual, sound analysis software works best with music where only one sound happens at a time. Here it is illustrated with our didjeridu example again.