PART IV: THE SOUND

Chapter 7: Introduction to the analysis of sound

The complexity of sound

Notation versus sound

Qualities of musical sound

Sound, body movement, and the club environment

Methodological challenges for the study of sound

The use of the term “sound” in musical contexts

Three types of sound events

The description of sound as timbre

The three domains of sound

The visual representation of sound

Summary

Chapter 8: Analysis of sound in electronic dance music

Analysis of the poumtchak sounds

The downbeat: pitch movements in bass drum sounds

The upbeat

The backbeat

Analysis of synthesizer sounds

Pitch movement in synthesizer sounds

Analysis of sound in effect processing

The gradual opening low-pass filter

Analysis of sound in the total mix

Compression

Summary

 

 

PART IV: THE SOUND

 

In the production of electronic dance music, a focus on sound is usually part of the process from start to end. Some producers may finish rhythmic, melodic, and harmonic structures for the entire track and then begin to exchange specific sounds and apply effects; others focus incessantly on these qualities along the way, mixing shorter segments ahead of the whole. Regardless, a finishing stage with a final mixing and mastering of the music is always required, and here the sound comes to the fore. Compression, reverb, stereo panning, and equalizing comprise some of the processes involved in controlling how the sounds work together. Thus, I have chosen to treat sound separately as the final part of this study. 

 

I will focus especially on those aspects of sound that might be related to notions of movement. In particular, I will explore how pitch movements in sound relate to body movements, and how different sounds interact in the experience of a groove. I will focus on sound on both a micro level, studying the various characteristics of specific sounds, and a macro level, looking at the possible outcomes of mixing processes and sounds effects in relation to longer segments of a given track.

 

Before I present my analytical material in chapter 8, I will discuss certain central problems and introduce some methodological tools in relation to my analytical approach in chapter 7.

 

 

Chapter 7:

Introduction to the analysis of sound

 

To fully examine the correspondences between rhythm in music and body movements, the actual sound of these rhythmic events must be addressed. But what challenges arise when sound is examined? What aspects or qualities of it are most significant? What methods should be used in an analysis?

 

Rolf Inge GodŅy describes the thorny problems related to the study of sound: “Questions of sonorous qualities are difficult because they do not fit well with the symbol-oriented paradigm of our musical culture (such as discrete pitch and durations), as sonorous qualities are highly multidimensional (evolving spectra, transients, etc.) and also rely heavily on introspective reports (i.e., pose difficult methodological problems for any experimental approach).”[1] In this introduction to my analyses of sound in electronic dance music, I will briefly discuss the problems presented here by GodŅy – the complexity of sound and the methodological challenges it represents – in light of my pursuit of the correspondences between music and corporeality.

 

The complexity of sound

 

notation versus sound

 

In musicology, the study of the sonorous qualities of music has seldom been prioritized. Analysis within the Western classical music tradition has typically favoured musical scores; as Richard Middleton points out, “Musicological methods tend to foreground those musical parameters which can be easily notated,”[2] including melodic, harmonic, and rhythmic structures. Middleton also observes that the musical score has often been equated with “the music” at the expense of the performance, so that the musicologist in fact views recordings simply “as a source for extracting the ‘fundamentals’ of the music, its ‘framework’, and condensing them into a score.”[3] This perspective is probably no longer as prevalent within musicology now as it was in 1990, when Middleton published his book, and several current research projects demonstrate an increasing interest in the performance and its recordings.[4] Nevertheless, Middleton’s words continue to ring true, especially with regard to the classical canon.

 

Recordings in popular music, on the other hand, are typically considered “the music,” not mutable versions or interpretations of an immutable score. In fact, most pop compositions only exist as recorded material, leaving little else with which the scholar can work. In the recording process, and especially through the development of more powerful recording technologies, qualities of sound that are not in fact easily notated become front and centre. Though many popular music scholars acknowledge this to be true, relatively few studies have broken ground in this field so far.[5]

 

Qualities of musical sound

 

There are significant qualities in music that cannot easily be notated.[6] These qualities have been trapped in the transitory nature of the performance, at least until developments in sound recording exposed them anew to scholars.

 

Musical sound consists of a complex blend of interacting qualities. Even in a single sound played by a single instrument, it can be difficult to separate and examine these various qualities; when several instruments are playing at once, the challenge increases. Pitch, timbre, loudness, and duration designate some of the qualities that are identifiable in a perceptual encounter with musical sound. They are subjective attributes that depend on measurable physical parameters – sound pressure, frequency, the spectrum, the envelope, and physical duration – and they relate to these physical parameters in a complex and interactive manner.[7] They can all be elusive, but timbre (tone quality/colour) is probably the most difficult to define. Pratt and Doak propose the following: “Timbre is that attribute of auditory sensation whereby a listener can judge that two sounds are dissimilar using any criteria other than pitch, loudness or duration.”[8] So timbre is what is left when the other attributes are excluded. Its physical parameters involve frequency, spectrum, and envelope, and the specific frequencies present at the sound’s attack the “transients” – are especially helpful for differentiating sounds.

 

However, pitch, loudness, and duration can also be tricky. Movements in pitch, such as intended or involuntary glissando/portamento or vibrations, can be difficult to decipher accurately, and these characteristics are often also affected by loudness (tremolo). In addition, amplification “is not just a question of volume,” states Theodore Gracyk: “High volume is employed to produce sound of a certain character, unique to electronically amplified music. There is a reciprocal relationship between volume and sound quality; increasing the amplitude of a sound wave alters its characteristic pattern and thus its timbre.”[9] While duration may boast the most straightforward relationship between its physical parameters and its perceived attributes, surrounding sounds in a mix (or one’s particular location while listening) can interrupt a reliable connection between the two. So how should sound be measured best? To be both true and relevant to a particular listener, sound measurements would have to be conducted uniquely at the exact time and place of the listening experience in question, because volume, speakers, the room, and myriad other factors combine to profoundly shape specific acoustic events.

 

Sound, body movement, and the club environment

 

The complexity of sound and the many quandaries related to its study demand a particular kind of focus that entails certain limitations. Body movement in relation to the poumtchak pattern will comprise the focal point of the analyses presented in the next chapter. All of the subjective attributes of sound (pitch, loudness, timbre, and duration) are addressed to some extent, but pitch movements within sounds prove richest in their implication. They relate most directly to issues of verticality in music and are therefore also intimately connected to my understanding of the effect of the poumtchak pattern.

 

It is particularly important to account for the limitations involved with measuring sounds when we analyze music that is intended to be played on large sound systems at clubs. My digital music analysis software measures sound relations as they exist on a recording – that is, before they produce actual changes in air pressure via audio speakers. The analysis of level of frequencies, decibels, and durations may only account for starting points in the experience of these sounds in their “real” environment. To comprehend the relationship between, for example, loudness and timbre, and the “behaviour” in a club of these sound qualities, an advanced understanding of and practical training in recognizing acoustical features is necessary. Nevertheless, pitch movement and other issues related to body movements are probably even more pronounced on a huge sound system, where the high volume directly contributes to their physical impact.

 

Methodological challenges for the study of sound

 

The study of sound within popular music studies is manifestly hampered by a lack of analytical methods that draw upon a standardized visual representational system, a common theoretical underpinning, or a consistent scientific vocabulary. The following discussion, then, deals primarily with terminology for and communication around sound.

 

The use of the term “sound” in musical contexts

 

A comprehensive definition of “sound” would have to include anything that is audible, and even in music, sounds that are not thought to be “musical” in a traditional sense may well be used. But pure sound does not generally connote “music” when detached from a musical context, so a definition of music as sound is not particularly instructive. There are in fact three ways in which the term sound is commonly used in musical contexts:[10]

 

1. “Sound” as a descriptor of a discrete event that originates from one definite source – for example, a bass drum sound, a single tone played on a piano, or a synthesizer (or sampler) sound whose length equals the amount of time a key is pressed. The term may be used rather neutrally to designate the specific event alone, or it can be intended to embrace its (sound) qualities. The latter use evokes the subjective attribute of timbre. Sound is often used as a direct substitute for “timbre” as well – when someone says, “I like the sound of that bass drum,” they are probably not referring to pitch, loudness, or duration.

 

2. “Sound” as a descriptor for certain audible features that in most cases are characteristic or representative of a substantial quantity or quality – a specific instrument (“a guitar sound” or “the 808 sound”) or effect (“a filtered vocal sound”), a production (“the Joshua Tree sound”), a group or artist (“the Kraftwerk sound”), a producer (“the Spector sound”) or record label (“the Motown sound”), a city (“the Seattle sound”), a genre (“the disco sound”), or a time period (“the sound of the eighties”). Timbre may also be designated here, but when a more complex sound source is involved, a variety of musical elements are often implicated by this use.[11]

 

3. “Sound” as a descriptor of a more neutral totality (regarding references to specific sound sources or qualities) comprising all of the musical elements in an event, including several instruments sounding together. This practice is related primarily to the mixing process.

 

The second and third uses of the term will vary according to whether characteristic elements of sounds or a more neutral totality are in question. In mixing processes, these uses will often merge during the process of achieving certain sound-related goals – such as when a mix is formed according to a characteristic “house music sound,” or when the totality of the mix is modulated by a specific filter effect.[12]

 

three types of sound events

 

The three descriptors presented above provide a point of departure for how I have grouped the sound events from various dance tracks that I have chosen to examine. In my analyses I address the following three types of sound events separately:

 

1. A discrete event that originates from one definite source with a certain defined temporal limit.

 

2. A characteristic or representative outcome of a certain application of effect processing within a certain amount of time on part or all of the mix.

 

3. The result of the combination of several sound sources within a certain temporal limit in defined sections or throughout the track.

 

In accordance with these three types, the following sound events will be discussed in the analyses in the next chapter: (1) bass drum, hi-hat, snare drum, and synthesizer sounds, (2) the gradual opening of a low-pass filter, and (3) mixing techniques related to the use of dynamic compression.

 

The Description of sound as timbre

 

As I have already mentioned, the most difficult qualities to describe in sound are primarily those related to timbre. Descriptions might range from playing techniques and mixing processes to physical features and acoustic proportions or even purely subjective interpretations of how sounds are perceived. Considerations of timbre usually include how sounds are produced (on, for example, an instrument) or perceived, and actual physical parameters are not often included.

 

William Moylan illustrates how descriptions related to the production of sounds can be problematic. “‘Violin-like’ to one person may actually be quite different to another person. One person’s reference experience of a ‘violin’ sound may be an historic instrument built by Stradivarius and performed by a leading artist at Carnegie Hall. Another person may use the sound of a Bluegrass fiddler, performing a locally crafted instrument in the open air, as their reference for defining the sound quality of a ‘violin’”[13] Descriptions of instruments, playing techniques, and other means of production may be highly accurate, but these things are only indirectly connected to our perceptual experiences.

 

The vocabularies created to describe the perception of sound (as timbre) are sprawling and often quite imaginative. Imprecise metaphors (“bright,” “warm,” “open,” and so on) are often used as well. Moylan elaborates:

 

Describing the characteristics of sound quality through associations with the other senses (through terminology such as “dark,” “crisp,” or “bright” sounds) is of little use in communicating precise and meaningful information about the sound source. “Bright” to one person may be associated with a narrow band of spectral activity around 15 kHz throughout the sound source’s duration. To another person the term may be associated with fast transient response in a broader frequency band around 8 kHz, and present only for the initial third of the sound’s duration. A third person might easily provide a different, yet an equally valid definition of “bright” within the context of the same sound.[14]

 

In a cultural community or context where timbre is often discussed (such as a recording studio), the use of metaphors or other descriptive approaches may in fact be quite nuanced and articulate. The problem arises when these descriptive approaches are transferred to communities outside this context, where an objective scientific vocabulary would have to be much more accurate.

 

A scientific vocabulary does exist, of course, in the study of acoustics, but it has limitations in its value to the broader discipline of musicology. To fully understand the objective physical parameters of sound, both a theoretical grounding in the subject and practical training in its application are probably required. Since most scholars in popular music studies are educated within a traditional, notation-oriented environment, this training is largely absent. But this may be changing.

 

The three domains of sound

 

Cornelia Fales discusses how qualities related to timbre present problems in any communication about sound. She groups the realizations of sounds into three domains according to their course from source to recipient: the productive, the acoustic, and the perceptual: “The productive domain focuses on the sound source and the physical motion that produces the sound; the acoustic domain concerns primarily the spectral composition and transmission of sound; and the perceptual domain consists of the perceived sensation resulting from acoustic sound.”[15] She argues that a “careful description of timbre seems to need the vocabulary of all three domains to capture the reality of tone quality.”[16] This is also true for all of the qualities of sound that are difficult to decipher and represent through traditional notation.

 

I will apply these domains in my analysis and discuss the sound events in relation to all three:

 

1. The productive domain: Descriptions of how sounds are produced, for example through certain playing techniques on a specific instrument or through the sound synthesis or parameter settings on a synthesizer. These descriptions are obviously especially useful for musicians who are familiar with these instruments and/or with basic sound synthesis. (In the analyses I describe and discuss how sounds presumably have been made on various instruments in question.)

 

2. The acoustic domain: Descriptions of physical parameters, including suppositions about how the sound will behave given the changes in air pressure of a specific listening environment. These descriptions will be most useful to the trained scientist, scholar, or sound engineer who understands musical acoustics. (In the analyses, I present sonograms displaying certain physical parameters of sounds as digital sound files; in relation to specific examples, I likewise discuss how the sounds might react to a club environment.)

 

3. The perceptive domain: Descriptions of the subjective experience of sounds, often particularly related to pitch, loudness, timbre, or duration. These descriptions will probably be most useful to experienced listeners of the music in question who, in addition, recognize the language that is being used. (In the analyses, I use subjective experience as a point of departure for my interpretations regarding body movement.)

 

The visual representation of sound

 

Another methodological challenge involves the visual representation of those sound qualities that are not accommodated by traditional notation. The various existing attempts to visualize a sound mix do not really solve any problems related to the representation of timbral qualities.[17] Other types of notation (guitar tablature, chord symbols, experimental or alternative notation) or representation (sonograms, waveforms, various sequencer windows) also contribute in their own unique ways, but no standardized alternative is currently at hand. Digital software programs for analyzing sound are probably the best tools at this point for introducing both methods of analysis and visual representations into much broader scholarly contexts. These programs can produce useful visualizations of the physical parameters of sound, but their results depend on the type of material and the qualities being explored.[18]

 

Fales uses sonograms to visualize certain features of sound, in addition to written descriptions from her three domains.[19] Such visualizations are useful for elucidating the relationship between physical parameters and perceived sound attributes. In the following chapter’s analyses of sound in electronic dance music, I will also accompany my written descriptions with sonograms. While they do not represent the left-right stereo relationship in any productive way, this is not particularly crucial to electronic dance music. Because this genre of music should sound the same wherever you are in a club, sounds are rarely panned hard to either channel.[20]

 

Finally, it is important to note that music will always be what we experience through our physical encounter with it, and even the best new methods of visual representation only function as a guide to what we hear.

 

Summary

 

I began this chapter with a consideration of the challenges arising when the actual sound of the musical events occupies an analysis. Traditional musicological analysis has engaged scores, and those musical parameters that scores reflect well have in turn been prioritized. The analysis of sound itself, however, is more relevant to popular music studies, whose subject generally lacks scores.

 

The complexity of sound represents one of the major problems related to sound analysis. Physical parameters interact to form the basis for subjectively perceived sound attributes: pitch, loudness, timbre, and duration. Timbre is probably the most difficult of these attributes to decipher, but pitch, loudness and duration also represent challenging quarries. In the analyses in the following chapter, my focus on body movement results in certain priorities. Pitch movement within sounds is the most consistent topic since it most directly addresses the same issues of musical verticality that arose in relation to the poumtchak pattern, but qualities linked to other sound attributes will also arise.

 

Terminology comprises one of the methodological challenges related to the study of sound. The application of the term “sound” in musical contexts is a point of departure for three types of sound events that will be examined in the next chapter: (1) sound as a discrete event; (2) the outcome of an effect processing; and (3) the outcome of a combination of several sound sources (the sound mix). Cornelia Fales groups the realization of sound into three domains, which I will apply as well in my analysis: (1) the productive domain; (2) the acoustic domain; and (3) the perceptive domain.

 

Another methodological challenge involves the visual representation of sound. I have chosen to use sonograms to accompany my written descriptions, as no standardized method presently exists.

 

 


Chapter 8:

Analysis of sound in electronic dance music

 

In this chapter I will examine correspondences between the actual sound of the rhythmic event and the body movements it evokes. As in my earlier analysis of rhythm, the poumtchak pattern will be central to both the structure of the chapter and the analytical work within it. The grouping of three types of sound events introduced in the previous chapter is also important to the structure here. I begin with the sounds of the poumtchak pattern, the bass drum and hi-hat, as representations of the first type, “sound as discrete events.” This type is also represented by the next topic, the sound on the “backbeat.” I then turn to synthesizer sounds that introduce variations to a vertical movement pattern in ways evocative of the complementary rhythmic patterns discussed in chapter 6. A specific filter effect (the gradual opening of a low-pass filter) represents the second type of sound event, “the outcome of a certain application of effect processing,” and finally, the third type, “the outcome of a combination of several sound sources,” is represented by mixing techniques related to the use of dynamic compression.

 

Verticality in music comprises a vital analytical premise in this chapter, based on the theories discussed in chapter 4, and pitch movement within sounds is the most consistent topic that will arise in these analyses.

 

As it was in my discussion of rhythm, my main focus here will be on British and French house music from the second half of the 1990s, with occasional examples from elsewhere. I also pursue historic perspectives on this topic, insofar as they manage to illuminate the musical effects of changes in the use of music equipment first described in chapter 2.

 

In the following analyses the manner in which the sounds are produced will also be of interest. This focus corresponds to the first of Cornelia Fales’s three domains of sound, presented in the previous chapter (the productive, the acoustic, and the perceptive). I will aim to bring all three domains into play in order to render a comprehensive description of the various sound events that are relevant here.

 

 

Analysis of the poumtchak sounds

 

THE DOWNBEAT: PITCH MOVEMENTS IN BASS DRUM SOUNDS

 

Based on my prior discussions of verticality in music, it seems likely that a descending pitch movement will facilitate or even activate a certain body movement in the same direction. Motor schemas contribute to transferrals from a descending pitch movement in the musical sound to an actual downward movement in space. It is also likely that a more pronounced movement in the music will suggest a more intense body movement. Bass drum sounds that are used to form a poumtchak pattern often involve such descending pitch movements, and they can have an effect on how various tracks are experienced.

 

Bass drum sounds in the 1970s

New methods of producing bass drum sounds have been gradually adapted in dance music production over the past few decades. An early disco production from the 1970s, to begin with, would most likely involve a bass drum sound that was recorded using a specific microphone onto a unique track on an analogue multitrack recorder. A descending pitch movement could be introduced to this sound by loosening a single tuning lug on an otherwise equally tuned batter drumhead, or by tuning the resonance head lower in overall pitch than the batter head. This was probably not a priority, however, among the numerous parameters for shaping an effective bass drum sound (width, size, depth, material of the drum, drumheads, pedal, microphone type and placement, studio equipment, acoustics, playing technique, and so on). A 1970s disco production typically aimed at producing a dense and defined sound that worked well alongside the bass guitar, as in the sonograms below. Both productions stand out with bass drum sounds that are more apparent in the mix than was customary in this decade.

 

LoveToLoveBassDr

CerroneLoveBassDr

Figure 8.1: Sonogram of bass drum sound at 01:39 from Donna Summer’s Love to Love You Baby (1975).

Figure 8.2: Sonogram of bass drum sound at 01:26 from Cerrone’s Love in C Minor (1976).

 

Both examples seem to involve pitch movement (note the darkest areas),[21] but it is not especially noticeable when we listen to them, and, compared to examples from the succeeding decades, the movement is in fact quite insignificant.

 

Drum machines of the early 1980s

The Roland drum machines from the early 1980s (the TR-808 and TR-909) have bass drum sounds that are produced in various ways through analogue synthesis.[22] However, one’s ability to control the sound parameters is limited relative to a conventional synthesizer. The TR-909, for example, has only four knobs with which to shape the bass drum sound: tune, attack, level, and decay. The tune knob on the TR-909 works within a limited range, and figure 8.3 displays sonograms of four different settings (0 = the knob turned all the way counterclockwise and 10 = the knob turned all the way clockwise).

 

 

909Tune-01il10

Figure 8.3: Sonogram of bass drum sounds from the Roland TR-909. Tune settings, from left to right: 0, 3, 7 and 10 (attack: 0, decay: 0).

 

As is evident here, the TR-909 can produce a fairly pronounced descending pitch movement (the darkest diagonal contour). This outcome is best heard (and seen) when the tune knob is turned all the way clockwise. Though the attack and decay controllers cannot shape this pitch movement further, they can emphasize or deemphasize it to a certain extent.[23]

On the TR-808 there are only three knobs: level, tone, and decay. The tone knob does not tune the oscillator but instead controls a low-pass filter that at different settings reduces the high frequencies.[24]

 

TR808stiplet

Figure 8.4: Sonogram of bass drum sounds from the Roland TR-808. Tone settings, from left to right: 0, 3, 7 and 10 (decay: 3). Stippled line at 1 kHz assists in visualizing high-frequency content.

 

On the sonogram, differences among the four tone settings are barely apparent in the presence of high-frequency content at the attack (the example to the very right has the most frequencies above 1 kHz). With tone settings turned all the way clockwise, the low-pass filter reduces less high frequencies, making the sound punchier. Still, the sense of descending pitch movement is not as pronounced with the TR-808 as it is with the TR-909.

 

Bilde 2

Figure 8.5: Sonograms of bass drum sounds from the Roland TR-808 to the left (tone: 3, decay: 3) and the TR-909 to the right (tune: 10, attack: 0, decay: 7).

 

On the sonograms, the differences between the two sounds are obvious. The attack of the TR-909 starts with a tone around 200 Hz that falls rapidly down to below 100 Hz, while the sound from the TR-808 has a less defined tone with the highest density of frequencies permanently below 200 Hz. While both sounds are defined and concise, the TR-909 sound has a more pronounced descending pitch movement, which, in accordance with the notion of an experienced verticality in music,[25] is likely to affect body movement more profoundly than the TR-808 sound.

 

Bass drum sounds of the 1990s (and beyond . . .)

During the 1990s the options for selecting bass drum sounds increased with the advent of various pre-recorded sound archives. Producers could merge electronically produced sounds with recorded acoustic bass drum sounds and customize sounds to fit the needs of their productions. Some producers even used specific bass drum sounds as part of their artistic signatures. The Israeli 1990s production team Astral Projection (Avi Nissim and Lior Perlmutter) cultivated a specific descending pitch movement in their bass drum sound in relation to the genre trance. In the track Dancing Galaxy (1997; tempo 139 bpm) the poumtchak pattern forms the basic beat. The bass drum sounds appear as diagonal lines in the lower part of the sonogram.

 

 

Bilde 4

Figure 8.6: Sonogram of excerpt (eight beat cycles) from Astral Projection’s Dancing Galaxy, 1:01–1:03.

 

The drum attack starts with a tone around 400 Hz that falls rapidly (over a period of one hundred milliseconds) to around 100 Hz. Though it does not resemble a traditional acoustic bass drum sound, its function is obviously the same and it has a definite concentration of frequencies below 100 Hz. In the mix the bass drum sound (and its pitch movement) is clearly evident. Since the pitch movement is more pronounced than the TR-909, it seems to evoke body movement even more strongly.

 

Bilde 3

Figure 8.7: Sonogram of bass drum sound from Astral Projection’s Dancing Galaxy.

 

By way of contrast, figure 8.8 displays a sonogram of a bass drum sound from the track What They Do (1996; tempo 93 bpm) by the American hip-hop band the Roots. Its concentration in the frequency area around 100 Hz and below continues (without any pitch movement) after the attack. There are also much higher frequencies at the attack (transients) in comparison to the Astral Projection example. This sound in fact resembles an acoustic bass drum sound and may have originated as one,[26] though it was probably shaped in numerous ways to suit the production.

 

Bilde 8

Figure 8.8: Sonogram of bass drum sound from the Roots, What They Do.

 

In British and French house music, both the TR-808 and the TR-909 were used extensively for bass drum sounds during the 1990s. But other alternatives such as samplers and harddisc recording became increasingly available during this period. The Daft Punk track PhŌnix (1996; tempo 127 bpm) starts with thirty-two successive bass drum sound events, and the descending pitch movement is evident both in the sonogram and to the ear. The drum sound starts with a tone at around 200 Hz that rapidly descends to below 100 Hz.

 

Bilde 4

Figure 8.9: Sonogram of a bass drum sound from Daft Punk’s PhŌnix.

 

In the sonogram below, I have collected three bass drum sounds from various Basement Jaxx releases, picking excerpts where the instrument particularly stands out.

 

Bilde 3

Figure 8.10: Sonograms of bass drum sounds from various Basement Jaxx releases: (left to right) Samba Magic (1995), Red Alert (Jaxx Club Mix) (1999), Where’s Your Head At (2001).

 

The bass drum sound in the example from 1995 is tuned higher than the two others, with a tone at around 250 Hz and a descent to below 100 Hz. The tones of the two other sounds both start (at around 200 Hz) and end slightly lower than the first, and they are also longer in duration. Their descending pitch movement is evident to the ear for all three but somewhat more apparent in the first.

 

Though there are numerous house music tracks without an apparent descending pitch movement in the bass drum, it is probably more common in the many subgenres of electronic dance music than it is in popular music in general. The sonogram below displays a bass drum sound from the original mix by the renowned American producer Timbaland (Timothy Mosley) (2001; tempo 121 bpm) and a remix by Basement Jaxx (2002; tempo 131 bpm) of the Missy Elliott track 4 My People; both tracks use the poumtchak pattern as the basic beat.

 

Bilde 5

Figure 8.11: Sonogram of bass drum sounds from Missy “Misdemeanor” Elliott: 4 My People. Left: Timbaland’s original mix from 2001. Right: Basement Jaxx’s remix  from 2002.

 

Differences concerning the descending pitch movement are quite evident in these sonograms. While Timbaland uses a rather short, dense bass drum sound without an audible pitch movement, Basement Jaxx uses a bass drum sound that recalls the previous examples. In the practice of remixing, such features attached to sounds may be part of the signature of certain producers/remixers.

 

Bass drum sounds and microrhythmic relationships

As I discussed in part III, deviations from a structured grid typically underpin the assumptions of most studies of groove and rhythm. A fully quantized track does not fit this picture. However, sounds interact in intricate manners. If a bass drum sound with a descending pitch movement is placed exactly on the downbeat, the pitch movement may in fact locate the lowest point of a vertical movement pattern just after the beat. A fully quantized hi-hat sound on the following upbeat will then be experienced as “early” in relation to this body movement, and this “early” realization can produce a “push” or “drive” forward by giving extra energy to the upward movement as it rushes to catch up. This shows that pulse sensation is not a stable phenomenon that is automatically transferable to discrete, quantifiable entries but a perceptual process connected to how the body specifically feels the pulse, which in turn depends on musical features such as sound parameters, sound combinations, and tempo.

 

Tempo

Generally a descending pitch movement in a bass drum sound should fit the tempo of the song (and the body movement that the tempo inspires). A trance track is usually faster (around 140 bpm) than a typical house track (120–135 bpm), while a hip-hop track may have a tempo below 100 bpm. At slower tempi, a pitch movement will have to be decelerated, which blurs the bass drum’s “punch” and can prolong the experience of a downward movement. Such bass drum sounds do not work in a house track with a tempo between 120 and 135 bpm, where the descending pitch movement has to be faster to supply that punch.

 

the upbeat

 

The hi-hat sound that generally complements the bass drum in the poumtchak pattern does not incorporate pitch movement and therefore merits less discussion in this regard. Developments in sound production concerning the hi-hat sound follow those of the bass drum sound, from a live drummer to drum machines and samplers with external MIDI sequencers and then to digital, computer-based audio and MIDI sequencers. Here the issues primarily concern duration, timbre/pitch, and the use of additional or alternative sounds in relation to the hi-hat.

 

Usually the same hi-hat sound is used from start to end in a track, and its timbre and duration are preserved. But every so often this will change, depending upon other features of the mix. Hi-hat sounds are also frequently modulated by filter effects, when these are used in the high-frequency areas of a given track.

 

Duration

In a section from 0:46 to 1:24 in Daft Punk’s PhŌnix (1996), the duration of the hi-hat sound gradually increases. This can be done by extending the MIDI entries in a sequencer or altering the settings of the sound’s “amp envelope” by, for example, slightly increasing the decay time while the sustain level is set to zero.[27] This is not particularly common but does serve to illustrate the significance of duration in relation to hi-hat sounds on the upbeat. Over the thirty-eight seconds of this build-up section, the hi-hat sound is extended five times, at every sixteenth beat-cycle (or every fourth 4/4 measure).

 

    

Figure 8.12: Sonograms of hi-hat sounds from Daft Punk’s PhŌnix at 0:46, 0:54, 1:01, 1:09, and 1:16.

 

The duration of the sound increases from about twenty-five milliseconds when it is introduced to about 160 milliseconds during the last sixteen beat-cycles of the period. If a certain pull upward is experienced from the hi-hat sound, it is possible that this pull is slightly intensified, or the peak position of the movement is extended, each time the sound’s duration is changed.

 

DaftPunkHH

Figure 8.13: Sonogram of Daft Punk’s PhŌnix, 0:51–0:55, covering the first change in duration, with suggested movement curve (events other than bass drum and hi-hat sounds are not taken into consideration).

 

When a sound like this hi-hat sound, with the same qualities of pitch, timbre, and loudness, is repeated, even subtle changes in duration will affect its salience. In relation to a movement curve, the sound with the extended duration may keep the listener/dancer at the peak of an upward movement a bit longer, in this way extending the upbeat and in turn driving the track forward via the succeeding quicker downward movement. The whole section that includes these five extensions of duration lasts for about thirty seconds, and these subtle changes may well produce a gradually intensified experience that is typical of the build-up sections of dance music tracks.

 

Timbre/pitch

Changes in timbre and/or pitch also comprise potential variations among hi-hat sounds. In the transition from measures 6 to 7 in the Basement Jaxx track Jump n’ Shout that was discussed in chapter 6, the hi-hat sound changes slightly in duration and loudness but more so in timbre/pitch – in the sonogram below, the four entries to the left reach to about 13 kHz, while the four to the right reach above 16 kHz.[28] 

 

HiHatBoosted

Figure 8.14: Sonogram of hi-hat sounds from Basement Jaxx’s Jump n’ Shout, 0:16–0:19.

 

These changes make the hi-hat sound more salient in this track, where it already cuts through a dense overall mix in a frequency range of its own.[29]

 

Additional vocal sounds

The last topic to be discussed in relation to the upbeat concerns other sounds that may alternate with, or more commonly, accompany the hi-hat sound. Samplers and sequencers make it possible to extract any sound from a musical phrase, copy it, and paste it in somewhere else in the track. These samples or sound bits might be either instrumental or vocal sounds. In the following examples I will try to illustrate how various vocal sounds might influence a vertical movement pattern. According to the theories concerning motor-mimetic processes presented in chapter 4, such sounds are particularly suited to evoking participatory movements from how we will produce the sounds ourselves.

 

My first example is from the track Five Fathoms (1999; tempo 122 bpm) by the British group Everything but the Girl (writer/producer/DJ Ben Watt and singer Tracey Thorn).

 

EBTG5F

Figure 8.15: Notational representation of Everything but the Girl’s Five Fathoms, 0:00–0:04.

 

 

    EBTG5F

Figure 8.16: Sonogram of Everything but the Girl’s Five Fathoms, 0:00–0:04, vocal sounds circled.

 

The vocal utterance on the second (and sixth) upbeat is in fact a short groan with a somewhat mechanical vowel sound. It is repeated throughout the track right after a downbeat that is emphasized by a pick-up note with a snare drum sound. The groan goes well together with head nodding or upper-body bouncing, since an upward movement supports the bodily actions needed for the utterance. The vocal sound seems to emphasize the specific upbeat when it is “performed” within the movement pattern.

 

A similar use of an utterance appears on the track Sucubz (2001; tempo 125 bpm) by the French producer Ark (Guillaume Berroyer).

Ark

Figure 8.17: Sonogram of Ark’s Sucubz, 3:28–3:30, placed in a grid with the vocal sound circled.

 

The last upbeat in a series of eight beat-cycles (two 4/4 measures) is followed by a short vocal “yeah” right before the succeeding downbeat. The event occurs late in relation to the upbeat (and closer to the following sixteenth), but is experienced as part of it. The duration of the utterance, its pitch, and its character match a vertical movement pattern, and performing it while moving seems not only to emphasize the upbeat but also to somewhat extend the corresponding movement, since the duration of the “yeah” is longer than anything else that happens on the upbeat. It could well introduce variation to the experience of a vertical movement pattern.

 

The last example is from the Missy Elliott track 4 My People, originally produced by Timbaland and remixed by Basement Jaxx. Both of these mixes have many vocal sounds (moans, “yeahs,” and so on) on both upbeats and downbeats. Short vocal sounds exactly on the upbeat are present in both of these versions.

 

Timba4MyPeople

BJ4MyPeople

Figure 8.18: Sonogram of Timbaland’s original mix of Missy Elliott’s 4 My People from 2001, 0:35–0:37, placed in a grid, with vocal sound circled.

Figure 8.19: Sonogram of Basement Jaxx’s remix of Missy Elliott’s 4 My People from 2002, 1:24–1:26, placed in a grid, with vocal sound circled.

 

The vocal sound here is a short, intense intake of breath. It occurs several times early in the tracks but not as regularly as in the previous examples. In both versions hi-hat sounds on the upbeat are also audible and in the Basement Jaxx remix a synthesizer joins the hi-hat at times. Like the previous examples in this section, the breathing here seems to match a vertical movement pattern perfectly, and it may bring extra intensity to the experience when “performed” by the listener/dancer.

 

These vocal sounds all bring an extra dimension of participation into the music that in different ways introduces variation and intensity to the experience of a vertical movement pattern.

 

the backbeat

 

Given its prominent position in the standard backbeat pattern, the snare drum is probably the most important drum sound in pop/rock. In electronic dance music, at least with regard to the poumtchak pattern, this is not the case, however. Very often a snare drum or handclap sound is part of the basic beat, but, as discussed in chapter 6,[30] this sonic presence is rather slight relative to the bass drum and hi-hat sound. The functional role of the backbeat here seems only to be the introduction of some variation to the basic beat and the movement pattern it activates.

 

There is no standard snare drum sound in electronic dance music, but sounds that accompany a four-to-the-floor bass drum pattern must be shaped to fit this task. Compared to other popular music genres, the snare drum sound will often be boosted in low frequency areas in particular. Handclap sounds often alternate with or even accompany the snare. Usually the sound on the backbeat (either a snare drum or a handclap) will demonstrate consistency in its duration, use of reverb, use of equalizers, and placement in the mix. These traits all contribute to the realization of a clearly defined groove.

 

Snare drum sounds of the 1970s

The snare drum sounds of a disco track and a rock track from the 1970s differ markedly in several ways.

 

ComparisonSnare

Figure 8.20: Sonogram of snare drum sounds from the disco track by Candi Staton, Young Hearts Run Free (1976), to the left, and the rock track by Led Zeppelin, D’Yer Mak’er (1973), to the right.

 

The two sounds differ less in timbre, than in duration and use of reverb. The disco track is much shorter and dryer. A more apparent boosting of higher frequencies at the attack (transients) is also visually present in the sonogram of the disco track (the darkest area of the sonogram has a higher vertical extension). The ideal snare of 1970s disco was a very muffled sound that ultimately derived from small, insulated drum rooms. Several productions from this time period have been criticized for being too dry and dull, but in relation to the dancefloor, such drum sounds are easiest to move to the forefront of the mix.

                                                           

The drum machines of the early 1980s

Both snare drum and handclap sounds were central to the Chicago house tracks, but the rock-derived emphasis on the backbeat gradually became less and less important there. These sounds were frequently combined with bass drum sounds, mainly as part of a different basic beat but also perhaps due to the fact that 1980s analogue drum machines could not match the rock backbeat snare drum sound.

 

HTZH:TR8081.jpg

HTZH:TR8082.jpg

Figure 8.21: Snare drum sounds from the Roland TR-808. Tune settings from left to right: 0, 3, 5, 7, and 10 (snappy: 0).

Figure 8.22: Snare drum sounds from the Roland TR-808. Snappy settings from left to right: 0, 3, 5, 7, and 10 (tune: 0).

 

The Roland TR-808 had three controllers for the snare drum: level, tune, and snappy. The tune controller could tune the low-frequency content of the drum within a limited range (see figure 8.21), while the snappy controller regulated the amount of white noise (see figure 8.22), thus imitating the sound from the snare wires. Still, the product was a long way from an actual acoustic snare drum sound.

 

The Roland TR-909 had a more advanced sound synthesis for the snare drum sound, with a tone controller in addition to the two controllers of the TR-808. It could therefore better recreate the short, dry disco snare drum sound and its sharp, definite attack. The handclap sounds on the two machines were quite similar and could not be modified in any way (except via the level control). Both the handclap and these snare drum sound variations from the TR-808 and TR-909 have appeared on numerous electronic dance music tracks.[31]

 

Snare drum sounds of the 1990s

During the 1990s, digital sounds on samplers and subsequently through harddisc recording introduced new possibilities, and better technology also offered more control over the shapes of these sounds.

 

RowWCirc

ShazzWCirc

Figure 8.23: Sonogram of Romanthony’s Down 4 You, 2:48–2:50, snare drum sounds circled.

Figure 8.24: Sonogram of Shazz’s Fallin’ In Love (PT. G Remix), 0:00–0:02, snare drum sounds circled.

 

The examples above illustrate two somewhat contradictory sounds on the backbeat of a poumtchak pattern, both in combination with a bass drum sound. The Romanthony track has a sound that resembles a handclap rather than a snare drum, but it is fuller and more defined than a standard handclap sound (for example, that of the TR-808 or TR-909). The sound seems to emphasize the downbeat but may not pull downward in the same way as the bass drum sound. Its effect on a vertical movement pattern, then, may be that every second downbeat is exaggerated but also halted earlier than the preceding and following downbeats (due to the high frequency content of the sound). The Shazz track uses a snare drum sound that blends in with the bass drum sound. The high frequencies seem to be attenuated, almost as if we are only hearing their reverb. The sound gives extra emphasis to the downbeat and in turn exaggerates this downward movement in a similar manner to the Hercules and Love Affair example discussed in chapter 6.[32]

 


Analysis of synthesizer sounds

 

Synthesizer sounds last as long as a key on the keyboard is pressed. Consequently, these events typically have considerably longer durations than the sounds treated above. This offers an opportunity for larger pitch contours and more obvious pitch movements and notions of verticality in the music.

 

Pitch movement in synthesizer sounds

 

The production of pitch movements

With acoustic instruments such as trombones, string instruments, or fretless bass guitars, it is easy to produce continuous pitch movements. Instruments where it is more problematic to slide from one note to the next (for example, pianos or guitars with frets) may also produce them to a lesser degree. Either way, a correspondence between the slide and body movement is likely, [33] whether as the result of conscious production goals or coincidence. The term “glissando” has typically referred to any pitch movement, while “portamento” generally implies vocals and string instruments (a stricter continuous pitch movement). The terms “portamento” or “glide” name this effect on synthesizers.[34]

 

Synthesizers of the 1970s

In this sonogram of Keith Emerson’s Moog sounds from 1972, the portamento effect is circled at the beginning of each tone.

 

Bilde 10

Figure 8.25: Sonogram of Emerson, Lake & Palmer’s The Curse of Baba Yaga, 03:14–03:16, with portamento effects circled.

 

The extremity or obviousness of the portamento or slide effect depends upon the pitch interval it covers (and its parameter setting). The pitch movement introducing the first tone in the sonogram is longer than those of the second or third tones because its interval is larger. In the sonogram below, Emerson alternates between intervals of one and two octaves.

 

Bilde 13

Figure 8.26: Sonogram of Emerson, Lake & Palmer’s Lucky Man (1970), 03:40–03:43.

 

The “pitch bend” effect also produces sliding pitch movements, and, contrary to the portamento/slide effect, it can be controlled independently while the musician is playing. Usually synthesizers are equipped with a wheel or joystick for controlling the pitch bend effect, so the keys can be played with the right hand while the left applies the pitch bend.

 

Bilde 15

Bilde 17

Figure 8.27: Sonogram of Parliament’s Flash Light (1977), 0:03–0:07.

Figure 8.28: Sonogram of Parliament’s Flash Light, 0:34–0:36.

 

In these two sonograms a pitch bend effect on the synthesizer sound is clearly visible as dark undulating lines whose relative inconsistency indicates that it was done manually while playing.

 

New production techniques of the 1980s

MIDI sequencers allowed for pitch bend effects to be edited or programmed directly into them. The computer-based sequencers (which arrived around 1987) had the ability to visually represent such effects, making them very accessible. Synthesizer sounds could also be programmed with modulators such as envelopes or low-frequency oscillators to further control the pitch in various ways. Thus entire patterns of pitch movement could be composed into a sound of a certain length. When memory storage became a common feature on synthesizers as well around 1980, sounds with programmed portamento settings and/or certain modulators for their pitch could be programmed or acquired as presets when buying the synthesizer (or later on from external sound programmers). Compared to the balancing act of playing keys and controlling pitch bend, of course, pre-programmed sounds were much easier to use.

 

An early example of this synthesizer sound is Shep Pettibone’s influential 1983 remix of First Choice’s Let No Man Put Asunder. During the last four minutes of the track, there is a short synth sound that is placed on the upbeat and then mixed in and out. In the sonogram below, the pitch movement of the sound is easy to see.

 

LetNoManCircl

Figure 8.29: Sonogram of Shep Pettibone’s remix of First Choice’s Let No Man Put Asunder, 6:19–6:21, placed in a grid with note values, with the synthesizer sound circled.

 

As demonstrated by the grid here, the sound plays alongside the hi-hat, shaping the tchak in the poumtchak pattern. Even though the sound operates in the upper-middle frequency area, it fits well with the hi-hat sound and contrasts with the echoed bass drum (also visible in the sonogram) and (through most of the track) the bassline. The sound’s pitch movement does not seem to affect body movement in a particular direction, but its bounciness seems to encourage general body bounciness as well. The event comes and goes, probably adding a degree of variation to the poumtchak pattern. In the 1999 cover version of the track by Mary J. Blige, an almost identical sound works in the same manner.

 

Bilde 22

Figure 8.30: Sonogram of Mary J. Blige’s Let No Man Put Asunder, 3:19–3:21.

 

There are several ways in which synthesizer sounds may produce notions of movement within a basic poumtchak pattern. As opposed to bass drum sounds, where the descending pitch movement fortifies a like direction in body movement, synthesizer sounds are less predictable. In some cases they may do the same, but they may also move in and out of this shared positioning, creating parallel but asynchronous patterns or even opposing the poumtchak pattern. This type of conflicting verticality from pitch movements probably contributes tension to a groove in a manner similar to the conflicting rhythm patterns discussed in chapter 6.[35]

 

In the following I will focus on this tension among different movement patterns as represented by the grooves of various electronic dance music tracks from the 1990s.

 

Various examples from electronic dance music

A synth riff I find remarkably powerful forms the core of the track Mentasm by the American DJ/producer Second Phase (Joey Beltram) from 1991. The riff lasts four beat-cycles (one 4/4 measure) and is repeated throughout most of the track. For the first two minutes, the sound is unchanged; severe sound modulation then occurs through the rest of the track. The riff consists of a few tones tied together by descending and ascending pitch movements.[36]

 

Beltram

Figure 8.31: Notational representation of synth riff from Second Phase/Joey Beltram’s Mentasm.

 

Stan Hawkins describes the riff as a “steaming, industrial jerky rupture,”[37] created on a Roland Alpha Juno synthesizer, which was first produced in 1986. This synthesizer has a digital oscillator capable of producing multiple waveforms, an analogue filter, and a single ADSR envelope that can simultaneously modulate the oscillator, filter, and/or amplifier.[38] It is therefore unlikely that the envelope modulates the pitch movements of this riff, which is more likely the product of a portamento effect.

 

Metasm1

Mentasm2

Figure 8.32: Sonogram of Second Phase/Joey Beltram’s Mentasm, placed in a grid with note values, 0:00–0:02.

 

Figure 8.33: Sonogram of Second Phase/Joey Beltram’s Mentasm, placed in a grid with note values, 0:08–0:10 (after the introduction of bass drum sounds on the downbeats).

 

These various pitch contours could activate corresponding body movements. In accordance with my discussions on musical verticality in chapter 4, these ascending and descending pitch movements can be experienced as notions of movement and further initiate a corresponding movement pattern. In the sonogram below I have attempted to draw a curve to illustrate a possible up-and-down movement pattern. The curve is shaped according to the tones in the riff, its dynamics, and its pitch movements. The descending B-flat to E interval seems to produce the most energy in the riff.

Metasm1Mov

Figure 8.34: Sonogram of Second Phase/Joey Beltram’s Mentasm, 0:00–0:02, with possible movement curve, placed in a grid with note values.

 

After sixteen beat-cycles (four 4/4 measures) with the riff playing separately, the bass drum enters on the downbeats. There is no upbeat hi-hat pattern, but the bass drum may activate a vertical movement pattern by itself. In the sonogram below, the two seemingly conflicting movement patterns are portrayed together.

 

Mentasm2lines

Figure 8.35: Sonogram of Second Phase/Joey Beltram’s Mentasm, 0:08–0:10, with possible movement curves, placed in a grid with note values.

 

Both patterns are equally emphasized and probably generate movements in a similar manner. It is therefore possible to switch from one to the other and thereby create individual movement combinations. Thus accentuated peaks (the B-flat to E interval), shifts in timing (or placement), and disturbances or tension points are produced by the interaction between the poumtchak pattern and the riff. The drawings in the two sonograms below illustrate possible movement patterns synthesized from the sounds.  

Mentasm3

Mentasm4

Figure 8.36: Sonogram of Second Phase/Joey Beltram’s Mentasm, 0:08–0:10 with possible movement curve as an interacting combination, placed in a grid with note values. The movement curve is dominated by the poumtchak pattern.

Figure 8.37: Sonogram of Second Phase/Joey Beltram’s Mentasm, 0:08–0:10 with possible movement curve as an interacting combination, placed in a grid with note values. The movement curve is dominated by the synth riff.

 

In both illustrations the placement of the peaks and troughs generally avoids exact note values in the grid. Variations in energy or emphasis are illustrated by the heights of the peaks and disturbances or tension points are breaks or irregularities on the curves. In this way we can visualize how individual experiences of a groove are produced through such interacting connections and patterns. When contradictory movements are introduced successfully, it seems to produce tensions in the muscles that in turn arouse excitement. It comes to no surprise, then, that this track achieved considerable success on the dancefloors of the 1990s.

 

Two other examples of interacting patterns appear on Daft Punk’s album Homework from 1996. Throughout the track Burnin there is a synthesizer sound with a pitch movement that careens between low- and high-frequency areas.

 

Bilde 1

Figure 8.38: Sonogram of sixteen beat-cycles (four 4/4 measures) from Daft Punk’s Burnin, 4:40–4:49.

 

The undulating structure made by the synthesizer sound on the sonogram resembles the lines I have used elsewhere to illustrate a vertical movement pattern. It is likely that this pattern activates even more dramatic movement in the dancer, because the ascending and descending movement here is continuous rather than repeatedly re-triggered by two contrasting sounds. The two patterns are not mutually reinforcing, however.

 

SynthSoundBurnin+PT

Figure 8.39: Sonogram of Daft Punk’s Burnin, 6:10–6:12, with a movement curve indicating a vertical movement pattern, placed in a grid with note values.

 

As is visually apparent in the sonogram, the pitch movements of the synthesizer sound do not correspond with the body movement pattern indicated by the poumtchak. This pitch movement only occurs two times during the four beat-cycles represented by the sonogram, and its realizations are in fact opposed to the poumtchak pattern – the peaks of the synth sound are placed on the second and fourth beat-cycles. So how do they interact? It appears that the first and third up-and-down movements are increased in both strength and duration, while the second and fourth are decreased in the same way.

 

SynthSoundBurnin+new

Figure 8.40: Sonogram of Daft Punk’s Burnin, 6:10–6:12, with a possible movement curve indicating an interacting combination, placed in a grid with note values.

 

In the sonogram above, I attempt to illustrate a possible movement curve that is influenced both by the poumtchak pattern (the movement curve with dotted lines) and the pitch movement in the synthesizer sound. These may of course be experienced in quite divergent ways. Still, it is important to recognize the tension, interaction, and variation that sounds with pitch movements can bring to a poumtchak groove.

 

On the track Rock’n’Roll from the same Daft Punk album, there is another synthesizer sound with a definite pitch movement.

 

DPRockNRoll

Figure 8.41: Sonogram of Daft Punk’s Rock’n Roll, 4:27–4:29, placed in a grid with note values.

 

The sound is faded in after thirty seconds and continues with constant modification until it is faded out thirty seconds before the end. It starts with a sustaining tone that changes into four undulating movements of increasing intensity. This course lasts four beat-cycles, and the final pitch movement ends in a new sustaining tone. The sonogram above is taken from a section where this sound occurs by itself and its pitch movements are visually quite apparent. For most of the track this sound is combined with a poumtchak pattern and a few other drum sounds (handclaps and snare drum sounds).

 

DPRockNRollMC

Figure 8.42: Sonogram of Daft Punk’s Rock’n Roll, 2:39–2:40, placed in a grid with note values, with one suggested movement curve following the pitch movements in the synthesizer sound and one following the poumtchak pattern (dotted line).

 

I have in the sonogram above illustrated a possible movement pattern according to the pitch movements in the synthesizer sound. The three last undulations form a movement structure that closely resembles a counterrhythmic pattern, and the listener/dancer’s experience of it may be somewhat similar to that discussed in chapter 6 regarding this topic.[39] It is again possible (and maybe more likely) for one to maintain the movements evoked by the poumtchak pattern and experience the counterrhythm of the synth riff as constructive tension or conflict within this pattern. It is also possible to switch back and forth between the two movement curves.

 

Generally, various synthesizer sounds with pitch movements appear to contribute tension to the poumtchak in much the same way as the complementary or counterrhythmic patterns that were discussed in chapter 6. This means that the dynamics of unique sounds in the rhythmic fabric might have as much of an affect on movement as the design of the rhythmic patterns themselves.

 

Analysis of sound in effect processing

 

In the analyses so far I have engaged sound as a series of “discrete events” – primarily drum sounds and synthesizer sounds. Now I will turn to the second type of sound events introduced in the preceding chapter: sound as “a certain application of effect processing.”

 

the gradual opening low-pass filter

 

Filtering effects in electronic dance music reach back to DJs like Nicky Siano, who in the 1970s started experimenting with the use of equalizers in New York clubs.[40] The removal (and gradual or sudden return) of either low or high frequencies has since become a standard part of a DJ’s repertoire. With the harddisc recording techniques and digital effects processing units of the 1990s, such effects have also become available at the production stage. The most common effect used in house music production involves the removal of high frequencies with a low-pass filter – this is often called the “underwater” effect or “that neighbour’s stereo through the wall trick.”[41] On production equipment these filters are often combined with a resonance or Q-point setting that boosts the frequency area in which the filter starts its attenuation. A typical build-up, then, might involve a gradual opening of a low-pass filter, often combined with a boost in the various relevant frequency areas. The following examples suggest ways in which this gradual opening of the filter may relate to movement.

 

Examples from Daft Punk’s Homework

On Daft Punk’s 1996 album Homework, seven tracks employ the effect in question, predominantly to modify certain instruments or sounds rather than the whole mix. I will first present three examples from this album and then discussing their impact on dancers.

 

Four times during the track Burnin (tempo 124 bpm), a combination of sounds (at first resembling bubbling liquid) starts out with only low frequencies and then gradually receives higher frequencies until it fills the spectrum. Particularly prominent here is a long sustaining tone accompanied by short rhythmic sixteenths.

 

DaftPunkBurninOpLF

Figure 8.43: Sonogram of Daft Punk’s Burnin, 4:26–4:58, with the effect of an opening low-pass filter circled.

In the excerpt in figure 8.43, the bassline is removed and the drum sounds attenuated until the low-pass filter is fully opened, while a synthesizer sound with an alternately ascending and descending pitch movement (seen as repeating lines in the sonogram) is kept in front.

 

HighFidCirc

Figure 8.44: Sonogram of Daft Punk’s High Fidelity, 4:38–5:16, with the effect of an opening low-pass filter circled.

The track High Fidelity (tempo 126 bpm) also has several filter effects through­out. In the passage in figure 8.26 the filter modifies all of the instruments/sounds except the hi-hat and snare drum. This excerpt is also a typical build-up section, with several instruments dropping out until the end of the passage.

 

The track Around the World (tempo 121 bpm) begins with a low-pass filter on the entire mix. After thirty-two beat-cycles (eight 4/4 measures) lasting a total of fifteen seconds, the filter starts opening, a process that also lasts thirty-two beat-cycles. It ends with an ascending sound effect introducing the main instrumental riff of the track.

 

AllAroundWCirc

Figure 8.45: Sonogram of Daft Punk’s Around the World, 0:00–0:33, with the effect of an opening low-pass filter circled.

 

How might this type of effect influence a vertical movement pattern? In this last example, low frequencies dominate the mix until the filter starts opening. Gradually the high frequency sounds become audible, and the alternation of low and high sounds becomes clearer. As the low-pass filter gradually opens, then, the pull upward by the increasingly obvious hi-hat sounds may intensify the movements. The same effect seems to be present in the other two examples as well, but the various unfiltered sounds, render it somewhat less profound. In the sonogram below I attempt to illustrate how a filter effect might function in relation to movement as part of a build-up section in a track.

 

AllAroundWMC

Figure 8.46: Sonogram of Daft Punk’s Around the World, 0:00–0:33, with suggested movement curve.

 

The initial rhythm pattern with only low frequencies only moderately activates a vertical movement pattern. As the filter starts opening, however, the movements slowly intensify as the higher frequencies are gradually exposed. The increased lengths of the movement curve indicate this intensification.

 

Moreover, several simultaneously occurring features may be involved in activating the intensification of this build-up section. Eric Clarke links the sound of a gradually opening low-pass filter in the Fatboy Slim (Norman Cook) track Build It Up, Tear It Down (0:28–0:55) to the perceptual effect “of a continuous movement towards a sound source that is first occluded.”[42] According to the ecological approach to perception, our knowledge of how sounds behave in the physical world is part of the process of music listening and can thus be a source for the experience of movement: “High frequencies are absorbed and dissipated in the environment more rapidly than low frequencies, leading to the characteristic ‘bass heavy’ quality of amplified music heard at a distance.”[43] The opening low-pass filter, then, will resemble the act of gradually approaching a sound source, which may contribute to the intensity of the build-up.

 

Egil Haga, in his Ph.D. dissertation on correspondences between music and body movement, identifies several features that he argues may contribute to higher levels of movement activation: greater densities of events, large pitch variations, higher volumes (loudness), distinct articulations, bright timbres, and horizontally spread and thick textures.[44] Several of the features (density, volume, bright timbres, thick textures) are to a certain extent increasingly present with the opening of a low-pass filter.

 

Finally, the familiarity of this low-pass filter effect and the expectation it therefore evokes for cultural insiders matters with regard to its impact on movement. The usual introduction (or return) of certain instruments, a specific, intensifying groove, or a structural climax at the point when the filter is fully open all work to satisfy dancers’ expectations. When one is moving to this music, one “holds back” and then slowly lets go, physically manifesting the musical tension and release that take place in the build-up sections.

 

Analysis of sound in the total mix

 

Lastly I will address sound as the “outcome of a combination of several sound sources” via various effects that engage the total mix of a given track.

 

compression

 

Loudness is an essential issue for the producers of dance music with regard to radio play, club play, and a track’s reproduction in other locations or situations as well. Any given track must seem at least as loud as the one it follows; a decline in energy is not tolerable. Compressors and limiters are dynamic processors that level out the volume of an instrument, which in turn makes it possible to boost it. Compression is often applied to different stages in the mixing process; to separate instruments, to instrumental groups, to the total mix, and finally to the mastering process. Dynamic processors can operate on specific frequency bands or on the total range of frequencies.[45]

 

A study of dance music tracks from the 1970s through the 1990s demonstrates the importance of compression and its enhancement. The amplitude representations below reflect a disco track, a Chicago house music track, and a British club/house music track, each separated by a decade.

 

Bilde 16

Bilde 17

Bilde 18

Figure 8.47: Amplitude repr. of Donna Summer’s Love to Love You Baby (1975).

Figure 8.48: Amplitude repr. of Steve “Silk” Hurley’s Jack Your Body (1985).

Figure 8.49: Amplitude repr. of Basement Jaxx’s Samba Magic (1995).

 

These amplitude representations indicate that the Basement Jaxx track from 1995 generates considerably more energy than the two others. These differences are also present on sonograms where the respective frequency bands are displayed.

 

Bilde 19

Figure 8.50: Sonogram of excerpts from (left to right) Donna Summer’s Love to Love You Baby, 0:32–0:56, Steve “Silk” Hurley’s Jack Your Body, 5:40–6:06, and Basement Jaxx’s Samba Magic, 1:36–2:01.

 

Again the Basement Jaxx track has more energy in all of the frequency bands. Improved control over the reverb of instruments has also allowed for greater volumes without blurring the mix. This development in the production of dance music has been essential to the music’s potential for activating body movements, with the enhanced punch from the bass drum being particularly important. 

 

Stereo panning

The diagrams in figure 8.51 represent the stereo panning of these tracks.

 

image012

Figure 8.51: Snapshot from a stereo analyzer (Elemental Audio’s Inspector XL) of downbeats from (left to right) Donna Summer’s Love to Love You Baby, Steve “Silk” Hurley’s Jack Your Body, and Basement Jaxx’s Samba Magic.

 

The Basement Jaxx track has almost all of its information centred, while the other two producers have created a much broader stereo image. Samba Magic is intended to be exclusively a dance music track, while the specifications of other media, like radio, probably informed the earlier productions. Sounds that are distributed among different channels can cause problems on a dancefloor that is organized with a left/right speaker system, because music can be missed. In many popular music productions, the drums are mixed according to their visual appearance on a stage with, for example, the hi-hat panned slightly to the right. In a dance music track, however, it is essential that the hi-hat sound is approximately the same wherever you are on the dancefloor, or else a vital component of the basic beat that drives body movement can be lost.

 

Ducking

One specific use of dynamic processors is called “ducking.”[46] Its main purpose is to emphasize a specific instrument by attenuating (ducking) others. In dance music this effect is often used to ensure that the bass drum is sufficiently present in the total mix, but it can also produce an interesting “pumping” sensation overall. In popular music production, “pumping” is the (usually) undesirable result of too much use of the compressor, but in house and trance music, where a steady four-to-the-floor bass drum controls the processor, it is often applied intentionally. If the threshold, attack, and release settings are right, the bass drum sound makes the processor reduce the gain automatically on the other sounds in the mix, but this reduction will be in effect only as long as the bass drum sound is present. Depending upon the attack and release settings, then, the gain may alternate (pump) at exactly the same rate as the poumtchak pattern.

 

In Daft Punk’s High Fidelity this dynamic effect is presumably in place.[47] The sonogram below represents the transition from a build-up section with a filter effect to the return of the bass drum sounds.

          

HighFidCompCirc

Figure 8.52: Sonogram of Daft Punk’s High Fidelity, 5:13–5:16, with downbeats circled and bass drum sounds indicated by triangles.

 

The downbeats prior to the transition (marked with small letters) have plenty of sound material above 200 Hz. A short vocal utterance can be seen in circle “a” and the beginning of a somewhat longer vocal phrase in circle “c.” Approximately the same sonic material can be seen vaguely within the circles indicating the four succeeding downbeats (marked with capital letters), but its volume has been drastically reduced. The beginnings of the vocal utterance (in circle “A”) and the vocal phrase (in circle “C”) are in fact barely visible though their continuations remain as they were before the transition.

 

A similar effect is present in the track Call on Me (2004; tempo 130 bpm) by the Swedish DJ/producer Eric Prydz. At the very start of the track, the effect is both audible and visible on the sonogram as the dynamic processor shapes a synthesizer sound with a sustaining tone.

 

Prydz

Figure 8.53: Sonogram of Eric Prydz’s Call on Me, 0:00–0:02.

 

At each bass drum entry, the volume of the synthesizer sound is reduced drastically, thereby giving the sound an oscillating character as it fills the spaces between the bass drum sounds. The processor does this throughout the track, resulting in quite audible ducking on the synthesizer sound.

 

This type of ducking would certainly appear to strengthen the contrast between a low-frequency sound and a high-frequency sound, thereby exaggerating the perceived verticality of the music.[48] When a dynamic processor is used in this manner, the bass drum sound dominates the mix and proclaims its “low” position in relation to movement. The experience of being pulled down by the bass drum will be enhanced when the high frequencies in the mix are attenuated. Moreover, the “pumping” effect compares closely to the alternation of movements activated by the poumtchak pattern and demonstrates how sound aspects are relevant in considerations of a tracks ability to evoke body movement.

 

Summary

 

My introductory analyses and discussions here concerned the sounds involved in the poumtchak pattern. The bass drum sound can incorporate a noticeable pitch movement, that in turn enhances its role in activating a downward body movement. An examination of production processes from the 1970s through the 1990s suggests that a downward pitch movement in bass drum sounds has become increasingly important to electronic dance music during this period. Various bass drum sounds, like variations in tempo, have come to characterize different subgenres as well.

 

The sounds on the upbeat (hi-hat) and backbeat (snare drum/handclap) do not typically incorporate pitch movements that might affect body movements. However, various aspects or realizations of upbeat sounds can emphasize or intensify particular positions in the movement patterns, based upon their duration or timbral quality, for example. Additional vocal sounds on upbeats can also introduce variation to a movement pattern, especially should listeners/dancers choose to participate actively in performing these sounds. Control of timbre and reverb has been essential in the production of snare drum sounds in dance music since the 1970s. Lastly, snare drum or handclap sounds on the backbeat may bring variation to the vertical movement pattern, depending primarily on their frequency content.

 

An outline of developments in the production of pitch movement on synthesizers from the 1970s through the 1990s demonstrated the new possibilities for control of these contours and intervals. Musical examples also revealed that pitch movements do not always correspond to the poumtchak pattern of the tracks and therefore affect movement in ways similar to complementary and counterrhythmic patterns, either bringing tension to or destabilizing a groove.

 

The use of a specific filter effect – the gradual opening of a low-pass filter – creates a steady increase in the presence of high-frequency sounds that can in turn intensify body movements. Increasingly higher density, greater volume, brighter timbre, thicker texture, the sensation of gradually approaching a sound source, and the culture-specific feeling of anticipation evoked by the low-pass filter effect probably all contribute to different degrees in the intensified movement experience of the build-up sections where the effect often occurs.

 

A comparison of tracks from the 1970s, 1980s, and 1990s supports the assumption that the use of compression has also been increasingly important to maximizing the energy level of dance tracks. This is probably due to the improved control over digital sound processing. In contrast, stereo panning appears seldom in dance mixes of the 1990s, probably to avoid the loss of any vital musical information when dancers are placed far from either left or right speakers. The specific ducking effect, where dynamic processors are used to deliberately reduce high frequencies on downbeats, can also influence the movement pattern. Given that perceived verticality in music can influence the experience of dancing, it appears that the lack of high frequencies may intensify movement on the downbeat.

 

Continue to Conclusion

Return to Index

Return to top

 

        

        

 

 



[1] GodŅy 2001:237.

[2] Middleton 1990:104. See also More 2001:34 and McClary & Walser 1990:282 for other perspectives on the prominence of traditional notation for music analysis.

[3] Middleton 1990:105.

[4] For example, CHARM (the Center for the History and Analysis of Recorded Music) is a research project led by Nicholas Cook that analyzes performance in European classical music through recordings; see webpage 7.1.

[5] Serge Lacasse’s study of voice manipulation (“vocal staging”) in recorded popular music is an exception; see Lacasse 2000. Projects associated with the Journal of the Art of Record Production and the Art of Record Production Conferences are also contributing to this topic; see webpage 7.2.

[6] See, for example, Roland Barthes’s observations in his article “The Grain of the Voice” (Barthes 1977). 

[7] Rossing et al. 2002:95 presents a table describing this complex relationship.

[8] Pratt & Doak 1976:317.

[9] Gracyk 1996:109.

[10] My interpretation here agrees with Ragnhild BrŅvig-Hanssen, who in her master’s thesis “Musikk og mediering” also specifies three categories for the use of “sound” as a concept in musical contexts: (1) Sound indicating a signature; (2) The total sound of a production; (3) The sound of individual sounds. My translation. In the original Norwegian: “(1) signaturangivende sound; (2) produksjonens helhetlige sound; og (3) individuelle lyders sound” (BrŅvig-Andersen 2007:7).

[11] Rob Bowman’s article “The Stax Sound: A Musicological Analysis” (1995) features examinations of ninety-five recordings from the Stax label that focus on various musical elements such as instrumentation, repertoire, structure, key, harmonic construction, timing, melody, ornamentation, and use of equipment (both instruments and recording equipment).

[12] See Brolinson & Larsen 1981:181–182 (in Swedish) and Michelsen 1997, chap. 5 (in Danish) for further discussion and another attempt to define “sound.”

[13] Moylan 2002:87–88.

[14] Moylan 2002:87. William Moylan’s book The Art of Recording aims to educate music producers and sound engineers using the author’s standards for understanding and differentiating parameters regarding acoustic relations. He offers several fruitful ways to represent various aspects of sound that could contribute to a more comprehensive methodology, but his hope for a truly objective description of sound may be unrealistic.

[15] Fales 2005:157; emphasis in the original.

[16] Ibid.

[17] See Danielsen 1998:283 and Moore 2001:121 for written descriptions of three-dimensional sound boxes where the axes represent frequency range/register, dynamics/depths, and left-right stereo mix/horizontal location. See Gibson 1997 (and webpage 7.3) and Michelsen 1997:137 for visual representations of similar images.

[18] Spectrograms were available prior to digital technology, but computer-based software programs have made such tools much more accessible.

[19] Fales 2005:167,171.

[20]Panned hard” means positioned fully to the left or the right channel in a stereo mix. See page 189.

[21] See my introduction to the use of sonograms on page 10.

[22] Using analogue synthesis, a producer can easily create a descending pitch movement in a drum sound with an inverted envelope generator whose medium short attack is set to modulate the pitch of the oscillator.

[23] The decay knob controls the AR (Attack-Release) envelope on a sawtooth-wave produced by an oscillator and shaped by a Waveshaper that cannot be controlled by any of the knobs. The attack knob controls the envelope on an additional sound produced by a noise generator (sort of a clicking sound). See Reid 2002a.

[24] The oscillator for the bass drum in the TR-808 does not produce a continuous sound (as most synthesizer oscillators do) – instead, it produces a sound that decays to silence without the need of any envelopes. The decay knob on the TR-808 controls a feedback loop that at various settings produces a shorter or longer sound. See Reid 2002a.

[25] See discussion on verticality in chapter 4 (page 134).

[26] The Roots are famous for playing hip-hop on conventional instruments (rather than DJ desks) and may want to give an impression of “liveness” also in their studio work. But even if parts of the track were played live in studio, they might have been looped; in addition, parts might have been replaced, mistakes might have been corrected, and so on.   

[27] The “amp envelope” regulates how the sound is amplified. The decay represents the second step in the ADSR (attack, decay, sustain, release) settings of an envelope, and it controls the duration between the sound’s peak level and its sustained level. The Roland TR-808 drum machine only had a decay button for the open, not the closed, hi-hat sound. 

[28] It is hard to determine the extent to which this change is caused by the introduction of a new hi-hat sound as opposed to the modulation of the original sound with certain new equalizer settings; therefore, it is also difficult to decide whether this change concerns primarily timbre or pitch. A traditional acoustic hi-hat sound cannot be tuned (except moderately through playing techniques), whereas a hi-hat sound on a sampler or a synthesizer often can be tuned up and down like any other sound.

[29] The term “staging” is used by several writers (Lacasse 2000, Moylan 2002, Zagorski-Thomas 2008) to indicate how sound is made to create a meaningful setting for the listeners. With the term “functional staging,” then, Zagorski-Thomas further applies the concept to “mixing techniques that spotlight certain functionally important musical features in ways that do not create specific musical meaning (such as highlighting an emotion or creating an illusory space) but which make the music function more efficiently” (Zagorski-Thomas 2008:204). He exemplifies this with production techniques in dance music that “strengthen attack transients and clarity” (205). This section from the Basement Jaxx track with a boosted hi-hat thus illustrates “functional staging” with regard to dancing. 

[30] See page 164.

[31] See Reid 2002b for a more detailed introduction to the production of snare drum sounds on the Roland TR-808 and TR-909.

[32] See page 165.

[33] The bassline on the verse of Lenny Williams’s You Got Me Running (1978), the background-vocals intro to the refrain on Thelma Houston’s Don’t Leave Me This Way (1976), the undulating string arrangements in the intro of Gloria Gaynor’s I’ve Got You Under My Skin (1976), or the ascending vocal intro on Chic’s Le Freak (1978) are examples from 1970’s productions with pitch movement effects that could impact body movements.

[34] Already in 1970 the Minimoog had a “glide” function that could be set to different levels using a knob on the front panel.

[35] See page 184. 

[36] A transcription of this riff is challenging, because very few tones exists as stable pitches. In addition to what I have transcribed, there is a pedal point on E-flat and a doubling of it two octaves lower. 

[37] Hawkins 2008:126.

[38] ADSR stands for attack, decay, sustain and release. A simple diagram of the synthesizer is available at webpage 8.1.

[39] See page 189.

[40] See Brewster & Broughton 2006:160–64.

[41] Preve 2006:95.

[42] Clarke 2005:81.

[43] Loc. cit.

[44] Haga 2008:183.

[45] Producers also use specific compressors or limiters to colour the sounds (changing its timbral content).

[46] Ducking can be achieved using regular compressors/limiters or through specific ducking applications. A side chain is normally used to feed the processor with an external signal (for example, the bass drum), which then controls the effect. See Izhaki 2008:270f and 376–381 for an introduction to ducking.

[47] The Norwegian sound producer Ulf Holand assumes the ducking here is applied to the total mix of the track with a bass drum sound extremely high in the mix beforehand (e-mail from Ulf Holand received 9. feb. 2010). 

[48] See discussion on verticality in chapter 4 (page 134).