Simon Zagorski-Thomas.

Introduction: Approach and Methodology

Putting aside the question of the geographical origins of my musical subject matter, I consider that my research falls within the territory of ethnomusicology simply because I’m using an anthropological methodology. This paper refers to African and Cuban recordings but the theory is more generally applicable to all forms of commercially recorded music. My fieldwork is, in large part, the twenty years I spent living and working amongst the ‘natives’ of the recording industry in London – learning their strange language and rituals and immersing myself in this alien culture almost to the point of being accepted as part of the tribe. This fieldwork as a sound engineer, studio owner, small time producer and composer has afforded me certain insights into the process of record production that have guided the more formal aspects of the research I’ve undertaken in the equally alien, tribal culture of academia. Although I’ve immersed myself in this culture of record production through participation, which ethnomusicologists would normally expect to lead to an ‘emic’ approach, my output is ‘etic’ in nature i.e. my descriptions and interpretations are not those used within the culture being studied.
Before I explain the concept of Functional Staging in more detail, I just want to briefly place this paper in a broader theoretical context. Probably the most widely studied aspect of record production up to now has been the analysis of the psychoacoustics of sound quality – visiting the back pages of the Audio Engineering Society’s journal should confirm that. Alongside this we find analyses that refer to the aesthetics of record production from a canonic perspective – studies of the creative processes that led a particular producer to a particular sound and how these have influenced those who followed. Studies of broader cultural reasons for the development of the sound of records are thinner on the ground and those that exist tend to focus on how the historical development of recording technology and techniques influenced the sound of record production. This paper is part of my attempt to search for further cultural explanations as to how the sound and processes of recorded music have developed both geographically and historically.

Functional Staging

William Moylan (2002. 1st edition 1992) developed the concept of staging as an educational tool as part of an aural training course for sound engineers during the 1980s and 1990s. The idea involves a conceptual or virtual stage in front of the listener relating to the way that sound engineers listen to a mix in the studio: equidistant from a pair of monitor speakers that are generally angled to form a triangle with the engineer as the third point. (Fig. 1).
Figure 1: standard placement of sound engineer in relation to monitor placement
 
Various techniques of technological mediation can be used to place the components of a piece of music at various places on this virtual stage (left and right. Near and far) to influence the listener’s perception of foreground and background and how various musical components might be conceptually grouped together.  Elements can be made to appear further to the left or right or closer or more distant than the actual speaker positions through the use of reverberation, panning, volume, equalisation and delay. This concept of staging can be further extended to include a vertical visualisation of frequency that adds a third, metaphorical dimension of low and high pitch to the two dimensional spatial staging that I have just described. Moylan and Serge Lacasse have developed the concept of staging to explain and describe aspects of record production from a creative and aesthetic point of view. My aim is to explain aspects of the staging process in relation to the function that the recording serves to its audience.
I’m going to start, therefore, with a short taxonomy of the functions to which recorded music can be put:
1.    Dance – playback in informal (party) or formal (club) situations. Production will ensure musical features important to facilitating the attentional synchronisation of dance gestures to musical gestures are highlighted.
2.    Focussed listening – playback for an individual (or small group) to listen attentively – mostly in the home but can be formalised (music society or acousmatic concert) or via headphones in other informal situations. Production will aim for clarity and stylistically appropriate proximity to suggest that the listener is a privileged (best position) witness.
3.    Performance atmosphere – playback to simulate or suggest the atmosphere of a ‘live’ performance. Production will reproduce, simulate or suggest acoustic properties associated with stylistically appropriate communal experience of a performance.
4.    Background – playback used for subliminal or peripheral creation of ambience where listeners attention is focused elsewhere. Production will aim to be smooth and without sudden dynamic or timbral variations.

These functions are not mutually exclusive and I will argue that different styles of music combine different aspects of these production approaches in different ways. Before I move on to look at some particular examples from African and Cuban recorded music, I’ll discuss a few wider generic musical descriptions and how these functional categories can be related to broad trends in record production.
One factor common to a wide variety of commercial recordings intended for dance  is that playback will be through a public address system in a large venue. The playback will thus entail the addition of substantial ambience from the dance venue itself as well as any ambience on the original recording. Reverberant spaces will blur the rhythmic characteristics of a piece of music by making the note onsets less distinct. These note onsets are the perceptual cues that we use to establish pulse and to synchronise dance gestures to musical sound. A characteristic of functional staging in recorded music intended for public dancing would therefore be to reduce the ambience on the recordings of the musical elements that are key to establishing the pulse of the music. In western popular music at the end of the 1960s and beginning of the 70s when clubs dedicated to dancing to recorded music started to become more popular, we see a divergence in drum sounds between dance music and rock music that seems to bear this out. (Example 1: Wonder 1973 and Led Zeppelin 1973)
At the same time in dance music, musical elements that are more concerned with generating the party atmosphere – most commonly vocals – are treated with reverb to suggest large scale communal activity (Example 2: KC & The Sunshine Band 1975) and contribute to the club ‘vibe’.
This use of production techniques to create the atmosphere of a large scale communal activity is used extensively in music that is designed for reproduction in a smaller home environment. (Example 3: Queen 1977 ). Rather than an accurate representation of the listening experience of a large concert hall though, the muddying influence of reverberating low frequency sound is usually avoided but the ‘fattening’ of the sound that this creates is often suggested through some sort of electronic or tape based compression of the low end. This gives some aspects of the perception of a large space without the loss of clarity that realistic reverberation would induce.
Many of the conventional techniques of multitrack recording and mixing can be related to this form of virtual staging – of generating psychoacoustic cues that are reminiscent of some features of a particular type of listening experience whilst avoiding other aspects which may have a negative impact on intelligibility or the musical meaning of a particular sonic feature. That, though, is a whole other paper.
Music intended for home listening through domestic hi-fi or personal stereo systems tend to involve some balance between the 2nd and 3rd functions in our list. I’ve already mentioned the approach that suggests or simulates a culturally appropriate communal listening experience and the other is aimed towards focused listening. This form of staging will often employ techniques that suggest intimacy and an individual approach – as if the performance is being whispered in your ear, and is solely for you. Close microphone placement, exaggeration of high frequency content and the relative high volume of dry signals in comparison to reverberation are all common techniques for suggesting proximity to the performer and when these are combined with low energy level, intimate performances the effect is even stronger.
In fact, these techniques have become so prevalent that in some styles of music they have become merged and confused with questions of recording quality – the closer they sound, the better the recording. This has also been combined with our continued exposure to unnaturally compressed bass frequencies to create expectations about the sonic characteristics of recorded music that constitute a culturally constructed perception of ‘good quality’ recording that extends well beyond questions of frequency and dynamic range.

Markets, Functionality and Modes of Consumption

I shall now present some examples from African and Cuban popular music recordings from the latter part of the 20th century to discuss ways that functional staging developed over time and how these are related to the availability and development of technology and to audience aesthetics.
One key issue in the evolution of and geographical differences in the sound of recorded music is the fact that economic development and the resulting technological infrastructure are crucial. Before the revolution in Cuba, its status as a kind of holiday playground for the USA meant that there were up to date recording facilities in this tiny geographical area that almost equalled those available throughout the whole of the African continent. Thus we get a 1957 recording by Chico O’Farrill like this:
(Example 4: O’Farrill 1957)
At this point in history we are before the advent of nightclubs using recorded music for public dancing and the production aesthetic is based around points 2 and 3 in out typology.
On the other hand, the recordings in the Congo and Senegal which were also being internationally distributed were created on relatively primative 2 track portable systems. This is the Congolese rumba group Ry-Co Jazz  who enjoyed considerable success throughout the 1960s both in Africa and the caribbean.  (Example 5: Ry-Co Jazz 1962)
As far as I can establish, these recordings made in Dakar, Senegal, consisted of a single microphone suspended above a live performance by the group in a relatively untreated studio space – a type 3 recording by default.
There’s also the same disparity between recordings made locally in areas with restricted access to technology and those where the artist records abroad. Witness the differences between the recordings made by Fela Kuti in Nigeria in 1968 and those made in Los Angeles in 1969. The LA recordings were initially only released in Nigeria so this wasn’t an example of an artist being transplanted into the western pop market but a recording made abroad for the African market. (Example 6: Fela Kuti 1968 – 9)
Both of these recordings are type 3 although the 1969 LA recording can be seen to have some type 2 characteristics such as the way that the solos are foregrounded. Although Fela Kuti was strongly influenced by the idea of the James Brown sound, these 1969 tracks have more in common with jazz – perhaps to do with the aesthetic of the American producer HB Barnham.
The next example is taken from a series of recordings made between 1970 and 78 by Ali Farka Toure for Radio Mali. The national radio stations, with a few exceptions such as EMI’s Johannesburg studio, tended to have the most advanced recording equipment in Africa and sometimes even doubled as commercial recording studios. These performances were part of a policy of promoting traditional culture and using the radio as an instrument of education and nationalistic exhortations to work hard and prosper. The mix is a type 2 recording that is particularly concerned with ensuring that the lyrics are prominent. (Example 7: Ali Farka Toure 1970 – 78)
This 1978 Cuban recording by Emiliano Salvador, the well known jazz pianist seems to have been quite heavily EQ’d for the CD release but also seems to have been influenced by the trend in jazz production of the time typified by the ECM sound of artists such as Jan Garbarek and Keith Jarrett – a live performance but one that utilised more ‘intrusive’ techniques that make the sound both cleaner and smoother – with more separation and fewer rough edges. A combination of types 1 and 2 with some characteristics of type 4.  (Example 8: Emiliano Salvador 1978)
The 1982 recording from Los Van Van’s seventh album in this next example represents a further step away from the recorded simulation of live sound. Some aspects of the percussion, especially the guiro, have been brought forward in the mix in a way that mirrors the treatment of percussion in western popular dance forms of the time. Although recordings like these weren’t being made for playing back to dancers in a club, the production aesthetic of type 1 recordings was creeping in and mixing with aspects of type 2, the prominence of the vocal for instance, and more stylised aspects of type 3, the fattened, compressed bass sound in conjunction with reverberation affecting only the higher frequency musical components of the track.  (Example 9: Los Van Van 1982)
Back in Africa, in this instance Ghana in 1985, economic factors are still of crucial importance. John Collins started the Bokoor Studios with a portastudio in 1982. This use of semi-professional equipment displays both the restrictions imposed by external economics and the way that these restrictions can lead to creative applications of technology. John Collins developed a technique based on the standard ‘bouncing’ process used with 4 track portastudios that involved recording the group more or less live minus the drums but including hand percussion and then overdubbing the drums afterwards. This created the clean percussion sound and his mixing emphasised the mid frequency drum sounds in contrast to the bass drum that is culturally consistent with West African traditional styles. The interesting aspect of this is that the market was for cassette distribution, John Collins tells me that there was no vinyl production in Ghana at the time. In this case the production is using local criteria for determining type 1 characteristics – the importance of high-mid frequency percussion – in contrast to the previous example of importing external type 1 characteristics that are not functionally relevant. (Example 10: Francis Kenya 1985)
The next example contrasts the production values of the 1996 Buena Vista Social Club album with the Afro-Cuban All Stars’ ‘A Todo Cuba Le Gusta’. The BVSC is using types 2, 3 and 4 in a mixture that creates the retro sound of a good quality 1950s recording while subtely replacing certain features of live recorded sound with technologically mediated substitutions such as the compressed bass end and lack of low frequency reverberation mentioned earlier. This is combined with overdubs and close microphone placement to give detail to the high frequency instrumentation that would be missing from a 50s recording but which has come to be associated with notions of high quality recording. The whole track is then tempered with the smoothing process of contemporary mastering compression that creates the possibility of using the album as the world music equivalent of ‘dinner jazz’ – it lends itself to being listened to intently or talked over in equal measure. The Afro Cuban All Stars’ album has taken the alternative route of mixing types 1 and 3 with a smattering of type 2 mediation. The percussion is close mic’d and much more ‘in your face’ to emphasise the dance gestures suggested by the performance whilst the vocals and horns contribute the suggestion of excitement through participation in a live performance – again though, the reverb and bass compression distance the mix from the reality of performance and position it as ‘quality’ recording – something to pay attention to. (Example 11: Buena Vista Social Club 1996 and Afro Cuban All Stars 1997)
My last example in this section contrasts the 1998 Papa Wemba album ‘Molokai’ with  Introducing Shiyani Ngcobo from 2004. Both albums follow the oft used African music formula of using British producers to tailor the sound for western audiences – Papa Wemba produced by John Leckie and Shiyani Ngcobo by Ben Mandelson. The Papa Wemba album was recorded in Paris and has very clean and smooth production values that relate to type 2 recording with the same hint of type 4 – the dinner jazz vibe – that was noted in relation to BVSC. The Shiyani Ngcobo album was recorded at Tropical Sweat Studios in Durban, although mixed in London, and uses many subtle features to combine types 2 and 3. Alongside the technological practice that we have mentioned before that is used to negotiate this balance between culturally constructed notions of quality and suggestions of live performance, we also have outdoor recording mixed with studio recording so that birdsong and ambient sound are recorded along with vocal tracks through recording in the studio garden and then mixed with cleaner rhythm tracks. (Example 12: Papa Wemba 1998 & Shiyani Ngcobo 2004)
An interesting aside at this point is that both the BVSC, through the album packaging and the film, and the Shiyani Ngcobo album, through describing the organic nature of the recording process in the sleeve notes, have used the recording methods as a way to sell the album’s authenticity. In comparison to the way that multi-track recording was seen as inauthentic performance to the point of cheating in jazz, the Shiyani Ngcobo sleeve notes sell the track laying process as a form of natural creative activity – demonstrating changing cultural attitudes to performance, composition and creativity.
The Cuban market became established as part of the ‘popular music’ canon through its penetration of the US market and despite the texturally different sound – most significantly the bass drum being absent or unimportant – it adapted the sound of its record production very early to the conventions of American popular music. Despite the fact that clubs that playback recorded Cuban dance music through PA systems are less firmly established in Cuba (and I’m ignoring disco derived forms and regetron for the purposes of this paper), I would suggest that structured dance forms with their historically European influence of formality, in comparison to African examples, put Cuban music in the arena of popular dance music production. The African market is more firmly in the more exotic ‘world’ category and has associations of communal performance that seem to have become more related to rock or even country record production conventions. This added exoticism in western perception has also led to a divergence in production aesthetics in African recordings intended for local rather than western markets that is less common or absent in Cuban recordings.

Authenticity, Performance and Record Production

The term authenticity would seem have little real meaning when applied in general to record production and musical performance. Authentic record production is any process that actually produces a record. It becomes a more thorny question when we start to apply it to qualified subsets of record production or performance – authentic folk music, an authentic 1980s Ghanaian electric highlife production sound. Was Dylan a folk musician before he went electric and a rock musician after? It’s interesting how certain sectors of the world music market have developed a similar preference for acoustic instruments as more authentic. Both Youssou N’Dour’s ‘Egypt’ and Mori Kante’s ‘Sabou’ albums demonstrate a move from electric to acoustic instruments in recent years in response to this shift in audience aesthetics.
Of course, if the categories themselves are personal and cultural constructs then so too are the definitions of authenticity – my definition of what constitutes an authentic blues track may be similar to yours or quite radically different. In the same way, certain techniques and processes of record production may be considered authentic and acceptable in some styles of music and not in others – but again the definitions will be personal or subject to negotiation with groups of individuals. A particular practice may be considered creative activity by some and meddling by others.
Certainly, authenticity isn’t about realism as we can see from the techniques that I’ve discussed which simulate particular facets of concert hall acoustics whilst inhibiting others. A more accurate concert hall style recording of the Buena Vista Social Club would have less high frequency definition and the bassiness would come from increased low end reverberation rather than the compression of instruments recorded with close microphone placement.
Of course, the idea of what constitutes an authentic performance on a recording has changed radically in the past half century as well as being different across musical styles. From the period when both audiences and musicians in jazz and classical styles considered the idea of splicing together multiple takes to be ‘cheating’, I would argue that musicians have increasingly come to consider it a matter of professional pride to both understand and utilise a wide variety of manipulative techniques in order to achieve the best possible finished recording. The level of manipulation and editing that is considered stylistically appropriate still varies dramatically from opera to gangsta rap but it would seem to be as much about satisfying audience expectations of what an authentic performance sounds like as about the musician’s pride in being able to produce the right performance in a single take.

Modernity and Exoticism

One particular aspect of the authenticity question in world music that I touched on when talking about preferences for electric and acoustic instruments, is the tension between modernity on the one side and tradition and exoticism on the other. Through my personal experience of engineering and producing south and west African bands in the 80s and 90s and through interviews with musicians and producers, I have observed that ideas about what constitutes an authentic contemporary sound for a style of music can be radically different depending on the audience experience and cultural background of the musician or producer.
It would seem that to musicians involved in the north African Rai or the Congolese Soukous scenes, the sound of synthesised brass or strings denotes modernity and an authentic contemporary production – these are the sounds in common usage in recordings produced for the local market. The acoustic instruments that they have replaced are considered ‘old school’ in the same way that particular electronic drum sounds are reminiscent of 70s and 80s dance and pop styles to western musicians. The aim is to produce a music with production values that would situate it within the mainstream of commercial popular music.
As noted previously however, to a large section of the western world music audience, the sounds of electronics and synthesisers are not considered authentic. It is an important aspect of it that world music sounds different – and the exoticism of unfamiliar acoustic instruments and vocal timbres is key to that difference. In fact, albums such as Sacred Spirit with its use of native American voices and Deep Forest with its use of pygmy vocal samples can be seen as extreme examples of the ‘foreign’ voice as an exotic instrument timbre.
Whether we consider this trend in aesthetics to be fetishising the exotic or celebrating the differences between traditional musics is largely irrelevant to this argument and, indeed, introduces an unwelcome moralising aspect to the question. The issue is that differing cultural traditions have developed different ways to approach the same style of music and these approaches appear somehow ‘wrong’ or inappropriate to audience members from the alternate cultural background.
In a similar way, historically unfamiliar combinations of technologies can appear inept or naïve. Thus, the use of heavily processed drum samples or electronic drum machine sounds in early to mid 1980s South African Mbaqanga and artificially high levels of hall reverb on the female choir vocals were inappropriate in the western world music market but popular and desirable in South Africa. In South Africa they were perceived as modern and performed the same function as in western popular music: clarifying and defining the rhythmic pattern in comparison to a low quality acoustic drum recording. In most cases this aesthetic wasn’t combined with the thickening of the bass line with compression and EQ and sounded like an inexpert and unsophisticated use of the technology. It didn’t fit into our perceptual timeline of technological development in audio production. It was a hybrid of 1960s and 1980s production techniques. It’s interesting how the juxtaposition of 1980s drum machine sounds and artificial reverb with the relatively unprocessed 1960s audio quality of the live instrument and vocal recording sounds ‘wrong’. (Example 13: Shikisha. Date Unknown) The Ghanaian recordings with live drums that we heard earlier used very similar technology and yet seem to fit our expectations better.

Conclusion: The Wider View and a Musicology of Record Production

This paper has extended my previous work on functional staging by introducing a taxonomy of approaches that can be used to describe a particular example of record production. The four approaches I have hypothesised can be seen to combine and interact in a variety of ways and, I would argue, are applicable to any form of commercially recorded music. I see the concept of functional staging as an important descriptive tool in a musicology of record production but as one that interacts with a variety of other factors such as technology, economics, cultural conventions and a more traditional aesthetic approach.  

References:

Lacasse, Serge. 2000 Listen to my Voice: the evocative power of vocal staging in recorded rock music and other forms of vocal expression. PhD Thesis. University of Liverpool.
Moylan, William. 2nd edition 2002. The Art of Recording. Focal Press (1st edition 1992)

Audio Examples:

Example 1:     Stevie Wonder ‘Living For The city’ 1973 Motown Records
Led Zeppelin ‘The Song Remains The Same’ 1973 Atlantic Records
Example 3:    KC and the Sunshine Band ‘Get Down Tonight’ 1975 Rhino Records
Example 3:    Queen ‘We Will Rock You’ 1977 EMI Records
Example 4:    Chico O’Farrill ‘Descarga Numero Dos’ 1957 Egrem
Example 5:    Ry-Co Jazz  ‘Gariophona’ 1962 Vogue Records
Example 6:    Fela Kuti ‘Wayo’ 1968 EMI / Zonophone
Fela Kuti ‘Wayo’ 1969 EMI / Zonophone
Example 7:    Ali Farka Toure ‘Samarya’ 1970 – 78 from Radio Mali Sessions of the period. CD release by World Circuit
Example 8:    Emiliano Salvador ‘Nueva Vision’ 1978 EGREM
Example 9:    Los Van Van ‘Hoy se Cumplen Seis Semanas’ 1982 EGREM
Example 10:    Francis Kenya ‘Memia’ 1985 recorded in Bokoor Studios, Ghana by John             Collins (released by Naxos 2002)
Example 11:    Buena Vista Social Club ‘Chan Chan’ 1996 World Circuit Records - recorded at Egrem Studios, Havana (Ocean Way LA, Livingston Studios, London and The Bakery LA).
Afro-Cuban All Stars ‘A Todo Cuba Le Gusta’ 1997 World Circuit Records - recorded at Egrem Studios, Havana (mixed at Livingston Studios, London)
Example 12:    Papa Wemba ‘Sakana’ 1998 Real World Records - recorded at Studio Davout, Paris and produced by John Leckie
Shiyani Ngcobo ‘Senzeni’ 2004 World Music Netword - recorded at Tropical Sweat studios, Durban and mixed at Church Walk Studios, London. Produced by Ben Mandelson
Example 13:    Shikisha. Unknown title. 1983 - 88. Unreleased