20th Century Music Vol 4(2) 2008
The Musicology of Record Production
Simon Zagorski-Thomas
London College of Music, TVU
Abstract
This article proposes an outline structure for a musicology of record production. This is intended as a starting point for debate rather that a definitive statement. Whilst recognising that a musicology of record production cannot exist as an entity separate to other aspects of musical analysis, it is hoped that a coherent set of tools for analysing the production of recorded music will prove useful. Six broad headings are used:
1. the development of technology
2. staging
3. ergonomics and embodiment
4. training and practice
5. the negotiation between performance practice and recording practice
6. consumer influence
These headings are used to examine the scope of existing work in this field and, in some instances, to make suggestions for future investigation.
Introduction
In 2005 I spent several months planning and co-directing a UK conference on the Art of Record Production. This article is a reflection on the many ways in which my thoughts on the subject developed during this time. When I was writing the call for papers, I devised a set of six themes which I then sent to various scholars working in this field for their comments and feedback. The initial themes were:
1 Transparency and distortion in recording. What are the aims of recording techniques? How close can we get to a transparent reproduction of a musical event? Is clarity more important than realism (e.g. multiple close microphone placements, track laying and editing)? Why do we like some forms of distortion?
2 What are the perceived types of authenticity in recording practice? Do musicians, technicians, and audiences have different ideas about authenticity in differing musical cultures? How important is the concept of ‘a performance’ in the notion of authenticity?
3 Milestone changes in production techniques and technology. How much has recording technology and professional practice determined the sound of particular genres or geographically defined musical cultures? To what extent are the ‘innovation giants’ of record production merely the first people to use a particular technique on a recording that became famous?
4 Why are the norms of recording practice in various musical sectors so different? Has the variation in textural parameters in different musical styles and cultures (e.g. melody and harmony in classical piano music, additive rhythmic patterns in West African Yoruba percussion music, vocal timbre in the Blues) been significant in the different ways in which recording practice has developed?
5 What is the product? What is the art object? Is it true to say that remix culture has affected the conception of the recording as the definitive artefact in Popular Music? Likewise, has the existence of multiple recordings of western classical works affected the idea of the score as the art object? And what effect has the recording industry had on musical cultures that are based on improvisation and variation?
6 In what ways can the perception of the meaning of a sound be altered by technology? Do gestalt grouping theories of music perception explain audio processing techniques which alter the perceived clarity of a recording? Are theories of music as a metaphor for the embodied expression of emotions borne out by the way certain audio treatments emphasize or even alter the emotional content of a performance?
As a result of a suggestion by Nicholas Cook, the conference also included a strand sponsored by the AHRC-funded Centre for the History and Analysis of Recorded Music (CHARM) entitled ‘Towards a Musicology of Production’. In many ways this was a way of signalling ‘all of the above plus anything we may have overlooked’, but it also provided a forum for the discussion of methodology. Indeed, it was when Allan Moore submitted the title of his paper as ‘Beyond a Musicology of Production’ that I started to draw together the various strands of thought that have resulted in this article.
I shall lay out my thoughts on the development of a musicology of production under six headings, but by way of introduction I should like to address three particular issues. The first concerns the ontology of music and its place in this argument. A musicology of record production suggests a recorded product. Characterizing music in these terms might also suggest that the recorded product is an autographic object for study in itself rather than an allographic instance of a work that should be considered without reference to the localized specifics of the recording. This is, of course, the conventional approach of both textual and contextual analysis in popular music studies. There are also areas of musicology, notably the study of western art music of the past three hundred years, where recordings are considered allographic instances of a work. This is equally true of the majority of textual studies in ethnomusicology, with the proviso that the ‘work’ may be a more flexible example of a text; it may, for instance, be characterized as a one-off example of a folk form or style. Recent developments, notably the ‘New’ musicology, draw on the methodology of ethnographic study and cultural theory to contextualize the study of western music. In this approach the study of record production techniques can be seen as part of a wider investigation of twentieth-century performance practice and its relationship to the production and consumption of music in the broader sense. A musicology of record production is thus not tied to the concept of the recording as an autographic work; it may just as easily be concerned with the contextualizing of allographic instances of a work in the same way that the study of performance practice can be.
The second issue relates specifically to Moore’s paper at the Art of Record Production Conference. In this he argues that ‘we can’t afford a musicology of production any more than we can afford a separately located musicology of the voice’. I agree entirely that record production cannot be studied in isolation from the music being produced. The creation of meaning in the recording process is a single part of the wider process of creating musical meaning in general. In fact the vast majority of the meaning contributed to recorded music by technological mediation stems from the provision of culturally appropriate staging for the musical content. While I believe that a ‘musicology of production’ is meaningless unless it is seen as an interactive component within musicology as a whole, this series of tools would make a more coherent and rigorous contribution to musicology if collected under a single banner.
The third issue is another ontological question: what to include and what to exclude. A literal interpretation of the term ‘record-producing technology’ should include everything from the pencil used to scrawl the lyrics, through flutes and synthesizers, to recording devices and the CD manufacturing process. While Moore’s assertion that we should be addressing all aspects of the production process (composition, arranging, and performance, as well as technological mediation) has been accepted for musicology in general, if we are to create the aforementioned series of tools, we need to delineate our area of interest. At first glance it might seem straightforward to differentiate between instrument technology and recording technology. A number of writers, however, have pointed out that sampling technology in particular makes this distinction problematic. The areas of overlap would seem to stem from the fact that some technologies can perform multiple and different functions in the production process. While we could argue for differentiating on the basis of function rather than form, Albin Zak has drawn attention to the various ways in which the recording process has become intertwined with the composition process. Hard-disk recording in particular has led to a ‘cut-and-paste’ approach to the composition of recorded music that is closer to creating a collage of sample-like snippets than to ‘recording’ one or more performances. The use of sampling and the way that signal processing has become inextricably linked with the creation of instrument timbres mean that we are going to have to live with a fuzzy border between the realms of instrument technology and recording technology.
The issue of where the recording process ends and where the audio product manufacturing process takes over is less problematic but nevertheless poses a few questions. We should ensure that mastering is included in the recording process, as this involves a creative shaping of the finished sound. But what of the selection of product media? These questions go beyond the choice between, say, MP3, CD, and vinyl to include the quality of the pressing or the choice of data compression techniques. While these are still factors that will affect what the listener finally hears, we have left the realm of creating the ‘master’ behind and have moved into the question of reproduction. The final master can be distributed on MP3, CD, and vinyl, and each of these will produce a slightly different version of the original master recording.
These three introductory issues can be summarized thus:
1 The question of whether we consider a recording to be an autographic or allographic instance of a work doesn’t affect the development of a set of musicological tools for the study of record production (although it will affect to what extent we consider ourselves to be studying performance practice rather than an art ‘object’).
2 A musicology of record production is a subset of musicology in general and as such it must be considered to interact with other subsets, such as harmonic analysis, performance practice, and so on, in any attempt to describe meaning in music.
3 This interaction means that there will be no distinct boundaries to a musicology of record production, particularly large areas of overlap occurring with instrumentation and arrangement.
I shall now move on to suggest a tentative vision of how I believe a musicology of record production might be constructed.
The Development of Technology
Probably the most obvious factor contributing to the transformations in recorded sound over the years and in different geographical locations is the development and dissemination of recording and production technology. In the mid-1920s we hear the change from acoustic to electric disc recording, and after the Second World War we witness the widespread introduction of tape-based recording, which the Germans had developed in the 1930s. Although Alan Blumlein patented a stereo recording system in 1931 and multiple tracks were used in film recording earlier than in the music industry, it was not until 1958 that the first commercial stereo records were released. During the 1960s and early 1970s tape track numbers expanded to three, four, eight, sixteen, then twenty-four. Simultaneously there was a change from valve to solid-state electronics, followed by the introduction of digital recording around 1980 and the move from tape formats to hard-disk recording in the late 1990s. During all these periods there were developments in product design that had a profound impact on recorded sound, such as improvements in microphones, mixing consoles, or speaker design, or the development of noise gates, the digital delay, and tape-noise reduction systems.
Several authors cover various aspects of these developments, but less frequently considered is the way in which the sound of recordings changed as a result of technological innovations. Although this has been charted to a certain extent in technical papers for the Audio Engineering Society, the data is often in the form of numerical specifications and measurements. There have been some interesting discussions on the Intenet about why we find certain types of distortion attractive in recording. It seems that recording engineers make equipment choices that represent a practice-based manifestation of this phenomenon in all genres of music and from all musical cultures.
Changes in technology can impact on the character of recorded sound in a number of ways. For instance, a series of incremental changes in a particular type of product may generate a wider palette of sonic options, one example being the development of microphone technology or dynamic compressors over many years. There have also been modifications that cause a sudden improvement (or a sudden change) in an existing technology, such as the replacement of the tape delay by the digital delay. While not changing the nature of the effect, this altered not only the quality of the delay sound but also the level and nature of the control that could be exercised over it. There are examples too of the introduction of a new technique or procedure that changes the recording process in some way. Thus the advent of noise gates in the early 1970s had a significant impact on the amount of space in the sound of popular music, and hard-disk recording (a realistic option from the late 1990s) made cut-and-paste editing so easy that it changed conventional working practices and hence the shape and feel of recorded performance.
More broadly, there are other aspects of the application of technology that have different implications for recorded sound. Studio design, room size, and acoustic treatment have a significant impact in this respect. I have observed elsewhere that the difference in sound between American and British record productions in the late 1960s and early 1970s was due mainly to the larger size of ‘live’ rooms in Britain. The development of the acoustically dead Westlake style of studio design in the early 1970s eventually spread around the world and made a highly significant difference to the sound of recordings. The enormous cost of this kind of acoustic treatment raises the question of economics in general. The costs of setting up a studio to match the shifting capabilities of professional practice rose steadily throughout most of the twentieth century, but recent developments have taken us back almost to the point of Fred Gaisberg in 1910, in the sense that a relatively inexpensive and portable recording set-up that doesn’t require years of special training to operate can produce recordings whose sonic characteristics meet professional norms.
A further question to be addressed here is how evenly technology has been distributed around the world throughout the history of recording, and how variations in distribution have affected the approaches to, and the sound of, record production in different places at different times. While many of the major record companies disseminated their technology around the globe fairly evenly (EMI, for example, started studios in India, Australia, Africa, and South America in the early 1930s and generally kept them up to the same standard as Abbey Road in London), their studios were so large that most local people could not afford to hire them. Particular developments in local production facilities around the world have led to the establishment of unique recording practices which can have a major impact on recorded music. Thus the importing into the Congo of a mono tape recorder by a Belgian musician in the 1950s had a notable effect on the spread of Congolese Rhumba; the establishment of the Shifty portable eight-track facility in South Africa helped develop black music recorded for a black audience; and British suppliers’ offloading (at a reduced price) portastudio technology to Nigeria that failed to sell in the European market had an impact on access to the modes of production in the West African music industry in the 1980s and 90s.
Staging
The concept of staging as a tool of analysis in record production comes from the work of William Moylan and Serge Lacasse but is also related to Trevor Wishart’s thoughts on ‘landscaping’ in electroacoustic composition. Lacasse refers specifically to the manipulation of the sound of the human voice, but his ideas are transferable to all recorded sound. The notion of staging refers to the treatment of sound in ways that add meaningful context for the listener. Perhaps the simplest example of this is the addition of ambience to suggest the sound source’s placement in physical space – a church as opposed to a bathroom, for instance. Directional, spatial, and distance cues in audio are more complex than this, however: the human perceptual system uses overall volume, the relative volume in different ears, differences in the arrival time of sound sources in each ear, the amount of high-frequency content, the relative volumes of direct and ambient sound, and the perceived performance intensity as suggested by the timbre. For example, two similar but different sounds might be heard at similar volume but arriving at the right ear slightly louder and slightly before arriving at the left. While this would obviously suggest the sound source’s orientation to the right, if one of the sounds had a greater high-frequency content than the other, we would be likely to perceive it as being closer, because high-frequency sound dissipates more rapidly over distance than low-frequency sound. Using equalization (EQ) to emphasize high frequencies in a recording is a standard technique to suggest intimacy and proximity.
Further complexity has been added in record production through the use of conflicting messages. If we listen to the recording of Whitney Houston belting out the final chorus of ‘I Will Always Love You’ and compare it with Jarvis Cocker’s vocal on Pulp’s ‘Common People’, we notice that the volume of the voice, the high-frequency content, and the level of room ambience are similar. The perceived performance intensities of the two vocal deliveries, on the other hand, are entirely different. Houston’s vocal timbre suggests high levels of energy being expended, and Cocker’s timbre gives the impression of a throwaway delivery and a world-weary lack of effort. The timbral clue of a quietly ‘spoken’ voice at a high volume level outranks other conflicting clues to stage Cocker’s vocal as intimately close and Houston’s as further away. An extra level of complexity is added in Houston’s case: although the intensity of the vocal can range from a virtual whisper at the start of the track to a powerful ‘roar’ at the end, the actual volume remains almost the same, controlled by a combination of compression and mix volume. The false impression we receive of how loud the vocal is at different points in the song, and our sense of how much energy is being expended in Houston’s singing, is shaped by the vocal timbre, and this overrides the opposing messages conveyed by the equality of volume throughout.
Perhaps, therefore, we can extend the concept of staging in a recording to include both the creation of meaning beyond physical placement within a perceived environment, and the way that timbral shaping can suggest emotional meaning. Here again, Lacasse’s work is paralleled in the field of electroacoustic composition. Denis Smalley discusses ways of generating meaning in electronic music by creating morphologies that suggest an action and an object; for example, a string vibrating when plucked, a human sobbing, or the smooth mechanical acceleration of a motor. Lacasse extends his concept of staging to include electronic treatments of sound that impose a timbral shape (or Smalley’s spectromorphology) on to recorded sound in ways that suggest the physical manifestation of human emotional activity. This is related very closely to theories of embodied cognition and perception, such as those proposed by George Lakoff and Mark Johnson and Antonio Damasio. Staging a guitar sound by adding overdrive or distortion creates a spectromorphology for that sound which is similar to the timbral shape of a shouting voice. By adding a certain pattern of both harmonic and non-harmonic overtones, the staging conveys meaning through relating the guitar sound to the type of emotional human states that we associate with shouting voices, for example aggression and anger.
The staging cues that we hear in a recording (irrespective of its type) are not exactly those we would hear in the actual situation. Even dummy-head binaural recordings, the method that most accurately captures sound as we hear it, doesn’t provide us with a completely accurate picture of an audio scene. We continually move our heads when listening, and the cross-referencing of our knowledge about our head orientation in conjunction with the natural staging information gives a much more vibrant and dynamic picture of an aural scene than any recording can. Most recordings we hear are very unnatural affairs compared with binaural recordings, and recording practice has developed a variety of unnatural staging devices to create different sources of interest in recorded sound; however, they are all grounded in the way we perceive sound in the ‘real world’.
Ergonomics and Embodiment
Several scholars have approached the question of how the ergonomics of new technological developments have shaped changing professional and creative practice. Zak has described the way in which multitrack recording techniques have changed the creative process for many musicians. Paul Simon’s Graceland album (1986) was partly written by recording extended ‘loops’ of basic sketches played by African musicians and then using the ‘feel’ to suggest structural editing and rewriting. It has also become common practice for writers to use sounds as the basis of an idea. Peter Gabriel’s creative process involves recording any experimentation that occurs in the studio and utilizing it as a springboard for the development of ideas, or storing it away for future use. These techniques evolved from changes in the technology rather than the technology being developed because there was a desire to make changes to the compositional process.
Paul Théberge has taken a somewhat different approach to this question. He charts the way in which keyboard manufacturers in the 1980s and 90s helped to alter the way that musicians viewed their instruments. The continued changes in the technology, and other factors to do with marketing, fostered the view that a keyboard was a consumable object that should be regularly upgraded. This can be identified as part of a more general trend whose effect is creeping into recording practice too. The market for plug-in effects and processors for the desktop sound-recording market has seen a similar move towards offering a large number of preset options (such as ‘rock hi-hat EQ’, ‘techno hi-hat EQ’ and so forth) rather than simply providing a variety of adjustable parameters. This is encouraging changes in working practice: rather than continually monitoring and tweaking the parameters of an effect or processor to fit the changing sound of a mix, an option is selected and maintained until it grates enough to be replaced by another.
The move from linear, tape-based recording practice to non-linear, hard-disk systems has also had a powerful effect not just on recording practice but also on the way that artists and producers conceptualize a piece and envisage the creative process. Many of the historical developments that Zak mentions in The Poetics of Rock can be seen as creating music through what might be described as ‘organic development’ (in terms of progressive growth), whereas the ‘cut-and-paste’ methods of desktop systems has encouraged composers to work in a modular fashion. It is becoming less common for any musician to play their part from beginning to end during the recording process. Non-linear practice often involves the producer or engineer aiming to record a ‘good chorus’ and a ‘good verse’ which is then copied and pasted to create the arrangement structure. This change in working practice has led to many composer-producers conceptualizing session musicians in the same way as they might envisage a sampler: as a sound source that generates modular units to be assembled and manipulated in the creative process. Sampling has to some extent altered the idea of composing to include collage and assemblage in ways that were previously perceived to be the domain of the DJ (the editor, the selector, and the impresario), driving changes in the way that non-linear recording is used. Indeed, the most successful software packages in this field have evolved out of MIDI sequencing software, thus further reinforcing the idea that non-linear recordings should be manipulated in the same way as MIDI sequences. This requires constant timing and playing with a click track to facilitate the editing process. Judging by the continued popularity of commercial music from the 1960s and 70s, it would seem that the rigid tempos imposed by click tracks are in many cases chosen for pragmatic rather than aesthetic reasons.
This brings us neatly to another way in which the ergonomics of music production technology have influenced creative practice. Contemporary non-linear recording software has added the visual dimension to editing sound in a way that simply wasn’t present in tape-based formats. Even during that period it was not unknown for engineers and producers to cover the VU meters on the mixing console with tape because they felt that they should make their judgements of sound quality purely from an aural perspective, without any visual influences. Recordists now have a graphic representation of every recorded sound wave available to them on screen, as well as a visual representation of the arrangement in the form of a block diagram showing which instruments have been recorded (or copied) at which points in the song. There is a potential rich vein of research in the study of how this (relatively) new visual aspect to the recording process has impacted on creative practice in the production process. This continuing process of atomizing the act of composition and record production and exposing every aspect of performance to closer and closer scrutiny has resulted in a clinical quest for technical perfection that often comes at the expense of aesthetic considerations. As soon as the technology to fix blemishes exists – through the use of compression to even out dynamics, for instance, or of the autotune to correct pitch inaccuracies – the pressure to utilize it is brought to bear; before long, since technical criticism is generally a lot easier than aesthetic criticism, its use will often become unquestioned standard practice. This pressure can also be applied if something looks ‘wrong’ on screen, even if the ‘flaw’ hasn’t actually been heard.
The impact on our perception and understanding of music of this quest for ‘perfection’ through edited performances and the mechanical repetition of sampled loops must obviously play a large part in any musicological approach that seeks to include an analysis of record production. There is neurological evidence that ‘a perceived beat is literally an imagined movement; it seems to involve the same neural facilities as motor activity, most notably motor-sequence planning. Hence, the act of listening to music involves the same mental processes that generate bodily motion.’ If, at a subconscious level, the perception of music involves hypothesizing what it would feel like to produce that sound, it would be useful for both music and musicology to study the grey area between edited performances that are perceived as possible and those that are perceived as impossible or unnatural. How much room for breath needs to be left in a spliced performance before we perceive it as an activity that is impossible to generate, and therefore artificial? How is it that we perceive the difference between a very accurately played repeated pattern and a sampled loop? There are, of course, many examples in recorded popular music in which room ambience or an unnatural truncation of a sound at a loop point will make the artificiality obvious, but there also seems to be another dimension to this that may be related to the impossibility of playing a perfect repeat. Theories of embodied cognition should be incorporated into the study of record production in the same way that they have contributed to other areas of musicology.
Another instance of technology and contemporary music developing in tandem is the way in which timbral innovation has become central to the composition process. The twentieth and twenty-first centuries can be characterized as a time in which timbral variation through extended and unorthodox playing techniques has become a central feature of both art music and popular music. In many instances this has been encouraged and extended by the ways in which technology has opened up further opportunities in the world of sound. It can be no coincidence that this trend has coincided with the development of what Eisenberg has called phonography, and the accompanying changes in listening practice. The ability to listen to the same performance many times allows the attention to focus on the minutiae of timbre, pitch, and phrasing that characterize these playing techniques. The study of performance practice is wholly reliant on recording technology to enable the measurement and analysis of these factors. I will not venture into the huge subject of electronic sound generation here, but the manipulation and processing techniques that have developed alongside the recording process, and the use of close microphone placement and multitrack recording, have all helped to focus attention on the grain of a sound.
Training and Practice
Edward Kealy’s article ‘From Craft to Art: the Case of Sound Mixers and Popular Music’, which concerns the development of the role of the sound engineer through the period from the 1950s to the 1970s, still seems to form the basis of academic study in the sound recording field. Kealy identified three modes of professional practice:
1 The Craft/Union Mode, in which sound recording was considered to be a craft, and training and organization were formulated and controlled by trade union bodies which functioned within the large corporations that owned the majority of recording studios.
2 The Entrepreneurial Mode, which saw the development of more adventurous techniques and less formal, though still ‘on the job’, training procedures. This mode was associated with the establishment of small, ‘independent’ studios in which the engineers were often the owners and were also, at least to some extent, self-taught and/or more prepared to experiment.
3 The Art Mode, developed out of the work of a few exceptional producers, such as Phil Spector and Joe Meek, who inspired both artists and engineers/producers to consider the application of technology as a creative act.
Kealy provides a good basic model, especially for the USA and the UK, but there are significant differences between even these two countries. The Entrepreneurial Mode was much less significant in the UK until after the establishment of the Art Mode; and there may also be good reason for arguing that large record companies awash with money from pop record sales in the late 1960s and early 70s contributed to the development of Art Mode techniques through the amount of freedom they gave musicians in the studio. As far as other parts of the world are concerned, many countries bypassed these models almost entirely and remained with the Craft/Union Mode until the introduction of cheap digital technology enabled them to move beyond Kealy’s model. Thomas Porcello has identified a significant change that began in the 1980s, namely the shift from apprentice- and practice-based learning to formalized post-secondary education. Although he is referring to the USA and Canada, the same trend can be seen in the UK and elsewhere in the proliferation of institutions such as the School of Audio Engineering around the world (some forty schools in over twenty countries) and the development of distance-learning schemes such as those offered by the Audio Institute of America. The Audio Engineering Society website also lists courses in thirty countries outside of North America and the EEC, and the number of degree-level courses with titles such as ‘Sound Engineering’, ‘Music Technology’, and ‘Sonic Arts’ has expanded exponentially since the mid-1990s.
The removal of barriers to entry by making equipment cheaper and simpler to use has also created another new mode of learning: the practical use of PC-based systems in conjunction with a plethora of semi-professional and amateur books, magazines, and, more recently, websites and internet discussion groups. These changes in the cost of entry have manifested themselves globally, albeit not equally, in different geographical and stylistic markets. Paul Greene and Thomas Porcello’s collection of articles Wired for Sound, concerning the use of music technology around the world, shows that this PC-based change has affected working methods, accessibility, and stylistic development across the whole range of musical cultures. Andy East has observed that Africa was often used as a ‘dumping ground’ for music technology products that weren’t selling in the developed markets in the 1980s. As we saw above, four-track portastudio tape recorder models that failed to sell in the UK ended up in Nigeria, and it would make an interesting ethnographic study to investigate whether specific items of technology influenced the sound of records produced in local markets. One distinctive difference in the development of record production outside popular music in North America and Europe (particularly the UK) is the relative absence of an Art Mode until the spread of PC-based systems and the dance-music-led cult of the producer. This seems to be true even of South Africa in the 1970s and 80s, when there was a small, powerful group of producers in the black music sector. Their power was not related to a particular ‘sound’ or to their artistic manipulation of recording technology, but was based on their ability as talent scouts and promoters.
The improved quality of ‘semi-pro’ equipment has prompted universities and other organizations offering recording courses to attempt to make a clear demarcation between what they teach and ‘bedroom’ recording, despite the fact that a significant proportion of commercial product is produced in cheap, PC-based project studios – the ‘bedrooms’ in question. Porcello has also referred to the way in which engineers seek to retain exclusivity in their profession through the development and use of specialized language. The vocabulary and correct use of technical language has the dual function of making communication between experts more accurate and efficient, as well as identifying the user as an expert and acting as a defining characteristic of the members of the expert community. He has also pointed to a sharp divide within industry professionals who trained via the ‘apprentice’ system and those who have progressed through formalized post-secondary education, an observation I am sure that anyone involved in the recording industry since about the mid-1990s would corroborate. I think that it is generally perceived on both sides of this divide that the ‘old school’ of work-based/experiential learning is a more ‘authentic’ method of acquiring knowledge. However, this may be the result of record companies selecting engineers and producers on the basis of track record rather than qualifications or other measures of perceived technical/artistic knowledge.
On a more general level, Timothy Taylor espouses a practice-based theory of technology. The design of the technology creates a structure that limits the influence of the recordist’s agency and will mostly determine the primary form of usage. However, the agents are also involved in ‘consuming’ or utilizing the technology. They ‘undermine, add to and modify those uses in a never-ending process’. Keep, approaching the subject from a background of contemporary electroacoustic composition, refers to the same phenomenon, which he describes as ‘creative abuse’. He provides a theoretical grounding for the extensive published work that relates to creative practice in the sound engineering and record production sphere. Authors like Bobby Owsinski offer privileged access to the ‘hints and tips’ of respected professionals. Indeed, this type of ’creative abuse’, in my experience, carries more professional or cultural capital than the technical knowledge that was previously more valued. Thus engineers would seem to set more store by the innovative use of technology (for example, misusing noise reduction technology to get a ‘warmer’ string sound, or using a speaker as large diaphragm transducer (microphone) to get a ‘fatter’ bass response) than by detailed technical knowledge (for example, an understanding of the history and physics of different stereo microphone placement techniques, or a knowledge of room acoustics). The engineers, perhaps unsurprisingly, will be assigning value to practice that favours agency on behalf of the user rather than determinism on behalf of the technology.
The Negotiation between Performance Practice and Recording Practice
Throughout my time as a sound engineer and record producer I was aware of the constant negotiation between performance practice and recording practice in decisions about the recording process. There seem to be several points of conflict between factors that musicians consider to be conducive to achieving a good performance and those that sound engineers regard as desirable or even necessary for obtaining a recording that meets the technical and aesthetic standards of contemporary recording practice. The underlying problem rests in the principally communal practice of musical performance and the desirability of isolating sound sources from each other in the recording process. The desirability of isolation stems from the use of multiple microphone techniques and the aesthetics of the ‘artificial’ staging of performances in a virtual environment.
Musicians, then, generally favour working in the same space, at the same time, with good lines of sight for communication and a live acoustic in the space so that they can clearly see and hear the rest of the ensemble and react to them. The technician’s desire for separation and isolation in the recording process has developed a practice that works in more or less direct opposition to these preferences. If musicians are playing in the same space and time on a recording, they are likely to be screened off from each other to reduce spillage from one microphone to another. This not only compromises their visual communication but will also reduce their ability to hear the rest of the ensemble clearly. The use of headphones obviously solves this problem but often at the expense of the musicians’ feeling of connection with one another. Screens are not always considered to provide a sufficient level of isolation, especially when loud instruments such as kit drums are involved. Musicians are therefore frequently placed in different rooms, which makes communication less direct, even when glass partitions are used. Multitrack tape recording has extended the possibilities for isolation by allowing musicians to record at different times: this removes not only the possibility of visual communication but also of two-way interaction in the performances. Negotiations between, on the one hand, ensuring the comfort of the musicians and creating the right atmosphere for them to stimulate the desired performance and, on the other, using recording techniques that provide separation are found in all genres of music and involve many different forms of compromise.
Frederick Moehn’s description of the changes in recording practice for the annual Sambas de enredo CD that accompanies the Rio de Janeiro Carnival offers a case in point. In 1999 the producers of the album decided to record the percussion tracks in the Company of Technicians Studio instead of the large circus tent that had been used previously. This involved separating out some of the musicians into isolation booths away from the main room and using headphones. Aside from saving money by employing fewer musicians, this change was instigated by the executive producer and the chief sound engineer ‘in order to arrive at a cleaner sound’. A contrasting solution is described by Beverley Diamond with regard to the Wallace family’s recording of their CD Tzo’kam in four different studios. In the first studio the drum was in a separate room from the vocals; in the second the singers were separated in different rooms and overdubs were added; in the third they were all in one room with separate microphones; and in the last they performed as they do live but with a single overhead microphone. Russell Wallace performed on and produced the album and describes the changes in terms of seeking a greater comfort-level with the performance arrangements, or what Diamond describes as the ‘social space of the studio’. The choices made are seen as being key to getting the right performances from the musicians and as being specific to this genre of music.
When I was recording jazz albums in London in the late 1980s and early 1990s I took part in many similar negotiations. The musicians had an idea of how they wanted the recording to sound, and typically that required a significant amount of separation; and yet they wanted to play in an environment that afforded as much interaction as possible. Factors such as the instrumentation, repertoire, and playing styles all contributed to the decision-making process. For example, the studio had a separate isolation booth for drums and a very small booth designed for vocals. The piano was in the main room and was usually screened off from the double bass in the same space. With quartets, the saxophone or trumpet would sometimes perform in the vocal booth and have artificial ambience added. Louder players, and often tenor sax rather than alto players, would frequently dislike the ‘feel’ of playing in the booth, and we would organize another screened-off area in the main room. Although this usually gave a less satisfactory audio quality, it was accepted as the ‘price’ of getting the right performance from the player. With quieter drummers the decision was regularly made to leave the door of the isolation booth partly open to improve the line of sight between the drummer and the bass player.
Paul Tingen’s description of the ways in which Daniel Lanois affected the working practices of U2 when producing The Unforgettable Fire (1984) and Achtung Baby (1991) shows that this negotiation is not confined to genres in which recording is generally restricted to capturing a single live performance. Although Lanois encouraged the band to record as an ensemble, he also extended to other areas of recording practice this desire for getting the right atmosphere so that the musicians could produce the desired performances. The chosen venues were not recording studios at all, but a castle and a rented house by the sea in Ireland into which mobile recording equipment was installed. Lanois was thus prepared from the outset to sacrifice audio quality and ignore the conventions of recording practice in order to achieve the right performances. This extended to the use of amplified monitoring instead of headphones, recording in the control room with the producer and the sound engineer rather than in a separate studio room, and generally allowing the recording process to take second place to the creative processes of composition and performance.
The issue of power relationships and studio politics raised here is not merely a question of aesthetics determining whose preferences prevail and to what extent. It was only when The Beatles had become a best-selling international commodity that they had enough clout to alter the recording practices at Abbey Road studios to suit their personal preferences. On the other hand, Louise Meintjes catalogues a sequence of events in the South African recording scene that demonstrates that musicians who, while associating ‘liveness’ in recordings with their African identity, have also had to compromise with engineers and producers to record what West Nkosi describes as ‘piece-piece – every individual plays alone’ overdubbed performances. The social dynamics of the recording environment offer many potentially fruitful avenues for future study, and some have been alluded to in the previous section, but this aspect of negotiating between performance practice and recording practice has a direct and palpable effect on the sonic qualities of the recorded output.
Consumer Influence
The final category that I am proposing examines the audience listening aesthetic and its influence on technology as a mediating force. It seems obvious to me that every style in recorded music is the result of a culturally constructed perception of what constitutes authentic recording practice; this perception is based on many different factors, including historical precedent, attitudes to different forms of technology, attitudes to performance practice, and the characterization of auteurship. Differing amounts and types of technological mediation will be considered to be authentic and acceptable by the different audiences affiliated to different styles of music. In a posting on the Musical Performance on Record discussion list in December 2004 Robert Philip evaluated two recordings of the same performance, one on contemporary digital equipment and the other on a 1940s system:
The modern stereo recording, compared with the mono 40s-style, put the audience in a much less comfortable relationship with the singer. Because of its clarity and impact – qualities characteristic of modern recordings – the singer seemed somewhat overbearing, right there just in front of us, and yet still singing as if we were a more distant audience in a concert hall. The ‘old’ recording seemed to place the singer at a more appropriate distance (though we were told that the microphones were very close together). The relationship seemed more ‘natural’. One could relax and enjoy the performance, instead of cowering from somebody who was singing insistently in one’s face.
The tape compression on the ‘old style’ recording altered the balance of low- to high-frequency sound in the same way that increased ambience would in a hall (low-frequency reverberation lasts longer than high-frequency), and thus the recording would reduce the impression of proximity by simulating one characteristic of hall ambience. Another characteristic of hall ambience is the reduction of clarity through the ‘smearing’ of one sound into the next by the reverberation. There seem to be a great many styles of recorded music that have developed the first characteristic without the second. Jeremy Wallach mentions that the preference in live dangdut recordings in Indonesia is still to use analog tape, just as (as Moehn confirms) it is the preferred medium for recording the Samba de Enredo albums, despite the fact that tape costs are much higher than those of digital recording. There is a similar practice in popular music of using electronic compression on bass parts and mixing them with drums that have real or artificial ambience added. The ‘smearing’ of the ambience is confined to non-pitched sounds and therefore doesn’t cause problems of pitch recognition, and yet a strong sense of live ambience is maintained.
This may require a slightly wider definition of the concept of staging that recognizes the practical use of staging as a way of highlighting textual elements that are important in defining the functional meaning of the music; for example, percussion might be highlighted to make dancing easier. This functionality is closely related to the norms of reception in the style of music. When the concert hall is accepted as the culturally ‘normal’ forum for a musical style, then we tend to see the use of compression and ambience in recording techniques. Natural ambience is often replaced by tape or electronic compression and EQ and a selective application of artificial ambience to higher-frequency and/or non-pitched musical components to avoid ‘muddiness’. Some of the characteristics of live sound are used to suggest ‘the real thing’, while unwanted side effects are eschewed. This allows the warmth or fullness of the concert experience but with an added clarity that is impossible to attain in the natural world.
An alternative aesthetic that has developed in contemporary recording practice is the exaggeration of intimacy. Through close microphone placement and the boosting of high-frequency content (which disperses more quickly than low-frequency content over distance), an impression of the performance being ‘up close and personal’ is created. Recording practices such as these are surely the result of an audience aesthetic that evolved out of changes in audio reproduction technology. In a process that began with the development of cheaper record players and the long-playing record, listening patterns moved from group listening (juke boxes and family gramophones) to bedroom listening and, more recently, to the personal stereo’s soundtrack-for-life in headphones. The development of audio staging techniques, which started with crooning and progressed to emphasizing every breath of a vocal performance and every scrape of finger movements on a guitar, has created the expectation that we will be addressed personally in certain styles of music. Moylan discusses this conscious manipulation of performance intensity and volume in contemporary record production in terms of the generation of meaning, both in the suggestion of intimacy and in the creation of foreground and background.
While these two forms of staging fall more naturally within Lacasse’s original definition of the term, the concept of functional staging may be useful in describing 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. Two widespread examples of this spring to mind. The first is the mixing of popular dance music during the past thirty years and more. Despite dance music being primarily aimed at an audience listening in group-based contexts such as clubs, the production aesthetic has moved away from a concert-based sound towards proximity and clarity in rhythm-section mixing and with the live aesthetic applied only to vocals and melodic instruments. It would seem that the ‘muddiness’ of concert-like recordings, which may give a desirable impression of power in rock music, is counterproductive when the function of the music is primarily to engender dance. Within a few short years dance music production has developed from the Motown and Stax sounds in the 1960s, which highlighted the rhythmic aspects of the track from within broadly the same live-sounding mix aesthetic as rock, to a mixing style that reflects the differing approaches of the audience: the functionality. Not only do we find drums getting louder and guitars getting quieter in dance music mixing (and the opposite in rock), we also find that production techniques that strengthen attack transients and clarity rather than power are also gaining favour. The music is shaped to make the rhythmic subtlety easier to perceive, the better to facilitate dancing (the function the music was created for).
The second example is the artificial boosting of vocal volume or clarity to make the lyrical content more audible. In fact this is so widespread that it interferes with other forms of technological mediation that are related to norms of reception. Thus, even in a style in which the mixing aesthetic in general is based on live performance, the vocal will often be made prominent to ensure that the lyric is intelligible, even if this reduces the impact of the instrumental performance. Other forms of functional staging might include, for example, the way that canto chico flamenco tracks were mixed in the 1970s and 80s. Drum kits were EQ’d very thinly in a manner suggestive of the palmas (hand claps) and pitos (finger snaps) of traditional flamenco. More than this, though, the low-frequency content of the kick drum and bass was reduced in a manner that seems to be making space for the full-frequency content of the acoustic guitar to shine through. The culturally determined ‘important’ instruments are given artificial prominence in the mix. Presumably it wouldn’t sound like flamenco to the target audience if the relative perceptual balance of musical components was upset. Percussion must be less prominent than the guitar, even when the percussion is transformed from the quiet finger snaps and hand claps to the louder kit drums, which would normally drown out an acoustic guitar.
Serge Lacasse has highlighted several examples of the way in which staging can be used creatively to subvert or exaggerate our perception of musical elements in a recording. He has concentrated on ways that suggestions of embodied meaning can be treated, such as using distortion on a vocal recording to suggest anger, pain, or fear. There are also many examples in various popular music styles of mixing techniques that subvert, exaggerate, and/or parody conventions that have arisen for functional reasons. The Flying Lizards’ 1979 single ‘Money’ is an extreme example, in which the mix has a deliberately thin and weak quality to support the mannered ineptitude and weakness of the recorded performances.
Conclusion
To some extent this article has been written to provide an Aunt Sally for interested academic parties to throw their sticks at. Any attempt to define the scope of a musicology of record production is bound to be incomplete. I’m hoping that this attempt will provoke discussion that might help to establish it as an important area of study. I hope I’ve combined a brief (and far from exhaustive) survey of some of the existing research that falls into my proposed categories with some suggestions for further work. Indeed, I am examining some of these areas, particularly in the last two categories, in my ongoing research.
I have concentrated on fairly distinct and delineated areas of study, but there are obviously many areas related to the production and consumption of recorded music that blur into other sectors of musicology. Performance practice, ontology, orchestration and arranging, compositional methodology, music perception and cognition, ethnography and cultural theory are all inextricably related to the study of the way in which recordings are produced (or perhaps vice versa), and it essential to remember that, as Allan Moore has suggested, the study of record production is a single component in the general study of music and has little independent meaning without reference to the whole.
Discography
The Flying Lizards. ‘Money’. 7-in. vinyl, Virgin VS276. 1979.
Houston, Whitney. ‘I Will Always Love You’, track 1 of The Bodyguard. LP, Arista 18699. 1992.
Pulp. ‘Common People’, track 3 of Different Class. CD, Island Records 524165 – 2. 1995.
Simon, Paul. Graceland. CD, Warner 7599254472. 1986.
U2 The Unforgettable Fire. CD, Island Records 206 530. 1984.
U2 Achtung Baby. CD, Island Records 510347 – 2. 1991.
Bibliography
Cunningham, Mark. Good Vibrations: a History of Record Production. London: Sanctuary, 1998.
Damasio, Antonio. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. London: Heinemann, 1999.
Daniel, Eric D., C. Denis Mee, and Mark H. Clark, eds. Magnetic Recording: the First 100 years. New York, IEEE Press, 1999.
Diamond, Beverley. ‘Media as Social Action: Native American Musicians in the Recording Studio’, in Wired For Sound, ed. Greene and Porcello. 118–37.
Eisenberg, Evan. The Recording Angel: Music, Records and Culture from Aristotle to Zappa, 2nd edn. New Haven, CT, and London: Yale University Press. 2005 (1st edition 1987)
Gracyk, Theodore. Rhythm and Noise: an Aesthetics of Rock. Durham, NC: Duke University Press, 1996.
Greene, Paul D., and Thomas Porcello, eds. Wired For Sound: Engineering and Technologies in Sonic Cultures. Middletown, CT: Wesleyan University Press, 2005.
Iyer, Vijay. ‘Microstructures of Feel, Macrostructures of Sound: Embodied Cognition in West African and African-American Musics’. PhD diss., University of California at Berkeley, 1998. Available at <http://cnmat.cnmat.berkeley.edu/People/Vijay/> (accessed 8 October 2007).
Kealy, Edward R. ‘From Craft to Art: the Case of Sound Mixers and Popular Music’, in On Record: Rock, Pop, and the Written Word, ed. Simon Frith and Andrew Goodwin. London: Routledge, 1990. 207–20.
Keep, Andy. ‘Does “Creative Abuse” Drive Developments in Record Production?’. Paper presented at the conference Art of Record Production, London College of Music, September 2005. http://www.artofrecordproduction.com/content/view/26/52/ (accessed 26 October 2007).
Lacasse, Serge. ‘Listen to my Voice: the Evocative Power of Vocal Staging in Recorded Rock Music and Other Forms of Vocal Expression’. PhD diss., University of Liverpool, 2000. Available at <http://www.mus.ulaval.ca/lacasse/> (accessed 8 October 2007).
Lakoff, George, and Mark Johnson. Philosophy in the Flesh: the Embodied Mind and its Challenge to Western Thought. New York: Basic Books, 1999.
McClary, Susan. Feminine Endings: Music, Gender, and Sexuality. Minneapolis: University of Minnesota Press, 1991.
Meintjes, Louise. Sound of Africa! Making Music Zulu in a South African Studio. Durham, NC: Duke University Press, 2003.
Moehn, Frederick J. ‘“The Disc Is Not the Avenue”: Schismogenetic Mimesis in Samba Recording’, in Greene and Porcello, Wired For Sound. 47–83.
Moore, Allan. ‘Beyond a Musicology of Production’. Paper presented at the conference Art of Record Production, London College of Music, September 2005 http://www.artofrecordproduction.com/content/view/26/52/ (accessed 26 October 2007).
Moylan, William. The Art of Recording: The Creative Resources of Music Production and Audio, 1st edn. New York: Van Nostrand Reinhold, 1992. 2nd edn, The Art of Recording: Understanding and Crafting the Mix. Amsterdam and London: Focal Press, 2002.
The Musical Performance on Record discussion list (MUS-PERF-REC@JISCMAIL.AC.UK). Thread entitled ‘Simulating 78s’, December 2004. Archives at <http://www.jiscmail.ac.uk/archives/mus-perf-rec.html>.
Owsinski, Bobby. The Mixing Engineer’s Handbook. Vallejo, CA: Mix Books, 1999.
Porcello, Thomas. ‘Speaking of Sound: Language and the Professionalization of Sound-Recording Engineers’. Social Studies of Science 34/5 (2004), 733–58.
Smalley, Denis. ‘Spectro-Morphology and Structuring Processes’, in The Language of Electroacoustic Music, ed. Simon Emmerson. London: Macmillan, 1986. 61–93.
Taylor, Timothy D. Strange Sounds: Music, Technology and Culture. New York: Routledge, 2001.
Théberge, Paul. Any Sound You Can Imagine: Making Music/Consuming Technology. Hanover, NH: Wesleyan University Press, 1997.
Tingen, Paul. ‘U2 & Robbie Adams: Recording in Mysterious Ways’. Sound on Sound Magazine (March 1994). Available at <http://www.soundonsound.com/sos/1994_articles/mar94/u2robbieadams.html> (accessed 8 October 2007).
Wallach, Jeremy. ‘Engineering Techno-Hybrid Grooves in Two Indonesian Sound Studios’, in Greene and Porcello, Wired For Sound. 138–55.
Wishart, Trevor. ‘Sound Symbols and Landscapes’, in The Language of Electroacoustic Music, ed. Simon Emmerson. London: Macmillan, 1986. 41–60.
Zagorski-Thomas, Simon. ‘The US vs. the UK Sound: Meaning in Music Production in the 1970s’. Paper presented at the conference Art of Record Production, London College of Music, September 2005 http://www.artofrecordproduction.com/content/view/26/52/ (accessed 26 October 2007).
Zak, Albin J. The Poetics of Rock: Cutting Tracks, Making Records. Berkeley, CA, and London: University of California Press, 2001.