Interaction of rhythm and sound in contemporary dance music

Anne Danielsen

University of Oslo

Studies in rhythm suggest that the design of musical rhythm at the micro-rhythmic level is of utmost importance for the stylistic and aesthetic features of rhythmic music. As regards research into the complexities of micro rhythm, however, much work up to this day has concentrated on the relation between norm and deviation, assuming that the main feature constituting what might be called the dynamic dimensions of a groove, that is, the aspect of a groove that makes a groove groove, is the presence of minor deviations from a presumed norm, for example a metric grid.
However, also music in which almost all events adjust to a metric grid, for example due to the use of programmed rhythms, does groove. As Hans T. Zeiner-Henriksen has demonstrated [see Hans’ paper at ARP ‘06], the sound of the different rhythmic events is crucial in this respect. In this paper the aim is to point to some additional aspects of the interaction of rhythm and sound in groove-directed music. This issue is among the primary research interests of the project Rhythm in the Age of Digital Reproduction [link to] at the University of Oslo, aiming at investigating changes in rhythm and sound in contemporary computer-based groove-oriented music, in which I take part.
In this paper, I will comment on how the choice of sound influences our perception of clashing rhythmic schemes that occur simultaneously, such as when a subdivision of two and a subdivision of three are used within the same groove, using Destiny's Child's ‘Nasty Girl’ (2001) as my point of departure. In keeping with this, I will also address the question of whether there is a transmission of information from the domain of frequency to the domain of time. Or put differently, I will discuss if sound as such - in the sense of timbral and dynamic content - may provide information about the placement in time of a rhythmic event, at a micro level, for example in relation to an internal basic pulse.   

The simultaneous occurrence of differing patterns of subdivision is a feature common to much groove-directed music. It may, however, take different forms. One example is to be found in a John Hiatt tune called ‘Thing Called Love’ (Bring the family, 1987). In this tune the guitar riff played by Ry Cooder has a shuffle feel to it, while Jim Keltner plays in accordance with a straight subdivision.
More often, however, the question of potentially conflicting patterns of subdivision announces itself in the form of ambivalence as to whether the subdivision is in two or three. This often means that the same rhythm may pull the overall feel of the groove in the direction of both swung notes and a straight subdivision. In my doctoral dissertation on the grooves of James Brown and Parliament, I found that a funk groove, for example, is characterized by inhabiting a certain borderland between duple and triple subdivision. (Cf. Anne Danielsen (2006). Presence and Pleasure. The funk grooves of James Brown and Parliament. Middletown, CT: Wesleyan University Press.)
My examples have so far been drawn from played music. In groove-oriented, computer based music the task of experimenting with different patterns of subdivision with different swing ratios, as well as combining different and potentially conflicting patterns, is quite easy. One example of how the increased possibilities of exploring such features of a groove can be made use of is the song 'Nasty Girl' from the album Survivor by Destiny’s child (2001). In this song, there is a certain ambiguity as regards subdivision that gives the groove a characteristic feel. For most fans, however, I do not think this ambiguity is experienced as a metric ambiguity. From a phenomenological perspective it should perhaps rather be characterized as a certain bodily feeling. From an analytical point of view, however, this feel may be conceptualized as a simultaneous presence of different patterns of subdivision, as is illustrated by the following slides.
Slide 1 (Subdivision in ‘Nasty Girl’, grid on straight 16ths)

As you can see, the rhythmic events fit well with a metric grid of  32nd notes or 16ths, depending on what one considers as the song's basic pulse. This means that the song has most likely been made by way of visually based programming. However, there are also rhythmic events in the programmed groove that fit in with a metric grid of triple subdivision, such as for example the pick-up to beat 4, or beat 2 in the case of a slow basic pulse.
Slide 2 (Subdivision in ‘Nasty Girl’, grid on triplets of eighths)

The most important aspect in establishing the hints of a shuffled feel in this song, however, is the vocals. One might also suggest that the vocals kind of mediate between duple and triple subdivision. It is neither duple nor triple, but rather inhabits the space in between.
It is my conviction that all this "fuzz" as regards subdivision in this song is an important element in creating the distinct rhythmic feel of the song. Moreover, I am equally convinced that if one replaced the sound of the vocal with a less flexible instrument, with a sharp, more definite attack, the effect would have been completely different. We probably accept this rhythmic clash because of the combination of sounds. Or put differently, it is the interaction of rhythm and sound that conveys the song's characteristic feel. Sound is thus an integrated part of the song's groove.

In other words, it seems like sound cannot be separated from the rhythmic features of a song. However, we might also ask, along these lines, whether there is a transmission of information the other way around: Does sound in fact provide crucial information to the listener about rhythmic events?
In my aforementioned study of 1970s funk grooves (cf. above), I put forward the idea that the placement of a rhythmic event in time may be recognized not only by referring it to the reference structure implied by the rhythmic fabric as a whole, but also through the timbral shaping of the event. As one of my examples, I discuss the fourth stroke in a riff played by the guitar in James Brown’s funk tune Sex Machine from 1971. This stroke is a typical example of the snappy phrasing that characterizes much of the performance of funk, where many ‘off-strokes’ are perceived as early and/or too short. I relate this way of rhythmic phrasing to James Brown’s expression ‘the downbeat in anticipation’, that is, when a note is being played on or almost on a beat but in a syncopated manner. Such playing may convey the impression of the beat being early, even though in a metronomic sense it is clearly within the tolerable margin of what would probably with a different phrasing be considered on the beat.
With closer inspection of a visual representation of the amplitude, it becomes clear that the fourth stroke of the guitar is played somewhere between the last 16th preceding the beat and the beat itself (exactly where is in fact very different to decide upon, due to the uncertainty regarding the correct location of the ‘norm’, that is, the pulse scheme that works as reference structure in this case). In other words, in this case a syncopation, an early beat, seems to be positioned in the space of the main beat. One might ask why this is so satisfying in a musical-aesthetical sense. One possible answer is the diverging interpretations this rhythmic event may evoke: the timbral shaping of the stroke provides the listener with information about the stroke’s location in time (at a sub-tactus or micro-rhythmic level), and the information given by this aspect of the rhythmic event is not fully congruent with the temporal location of the stroke in a metrical sense. In general, my study of 1970s funk grooves points towards the assumption that on a micro-rhythmic level, perceived durations, as well as perceived temporal placement in time, are dependent on timbral and dynamic features.
A familiar problematic is found in the way accents are used to group events in time. An accentuated beat has a different timbral and dynamic content than a non-accentuated, and when otherwise equal beats are played in succession, the first beat of a group (for example a measure) tend to be marked more strongly than the other beats of the group. As Sofia Dahl and also Carl Haakon Waadeland have demonstrated, the performance of an accent tends to have as a consequence that the beat becomes longer in time. This means that the beat following the accentuated stroke will be performed “too late”. (Sofia Dahl (2000) ‘The playing of an accent - Preliminary observations from temporal and kinematic analysis of percussionists.’ Journal of New Music Research 29(3); Carl-Haakon Waadeland (2006, forthcoming) ‘Strategies in empirical studies of swing grooves.’ Studia Musicologica Norvegica 32.) Moreover, experimental research also shows that when eliminating the difference in dynamic level between accentuated and non-accentuated beats in a sequence, timing information alone (in the form of Inter-Onset-Interval fluctuations) was enough to group the beats. (Dahl 2000: 231-32) There is, in other words, a link between dynamics on the one hand, and duration or placements of rhythmic events in time (at the level of micro rhythm) on the other.
An example from D’Angelo’s Voodoo album from 1999 may illustrate how the interaction of rhythm and sound may be used as an artistic device. On the track ‘Left and Right’ on this album there is a peculiar, almost seasick time-feel. One of the sources for the characteristic feel of this groove is the clash between the late realization of the syncopated figure played by the guitar just before beat two and the snare drum on beat two. Regarding placement in time, the syncopated note played by the guitar is very close to the second and fourth beats. Actually it is as close to the beat as it is to the syncopation on the preceding 16th note. However, in spite of its placement, very close to the downbeat, there is no doubt as to what rhythmic figure is being realized by the guitar. The second stroke of the guitar is commonly schematized as a syncopation. (In addition to my own interpretation of the groove, I have tested this out on groups of music students).
According to D’Angelo, every sound on this record is played, but the fact that it is extremely difficult, if not impossible, to play a guitar pattern this steadily unsteady for several minutes indicates a production process where the entire guitar track has been re-positioned, or more precisely, has been moved slightly to the right (according to how the music will appear when represented on a screen, for example as events in a sequencer program) after the recording. One hypothesis for explaining the peculiar feel of this groove would then be that there is a tension between the placement of the basic pulse suggested by the guitar, and the placement implied by the lower layers of the rhythmic fabric, namely bass and bass drum.
In keeping with this, one might also ask if the articulation of the closing of the figure played by the guitar would have been different if it were played where it actually sounds. Similar to the example in James Brown’s ‘Sex Machine’ discussed previously, there is, in other words, a discrepancy in ‘Left and Right’ between the timbral and dynamic shaping of the sound of the syncopated guitar and where it actually sounds. The sound of the guitar suggests a different placement of this rhythmic event in relation to the reference structure than what is actually the case.

Rhythm and sound have commonly been regarded as two separate domains in the analysis of music. However, as I have tried to demonstrate in this presentation, rhythm and sound are in fact interdependent. Moreover, recent developments in music technology seem to have encouraged a way of producing dance music that makes it particularly difficult to distinguish between the two.  Also recent research into the micro-rhythmic features of rhythm in grooves points in the direction of a mutually dependent relationship between rhythm and timbral and dynamic features of music.
Summing up, it seems that these aspects of the music interact to such an extent as to make it very difficult to distinguish between them, and, moreover, that the interaction of rhythm and sound becomes especially important to take into consideration when approaching the music at the level of micro-rhythm.