I am currently a postgraduate researcher, working towards an MA investigating innovative methods of environmental representation in film and film sound. I will be following this up with a PhD in sound texture synthesis. I have a first class BA from a leading UK university, experience working in a significant French research lab, and publications at IEEE conferences. I spend my free time creating installation art, sound design software and music, and also devouring a great quantity of books about anything and everything. My work is driven by a deep attention to detail, and I like to treat academic writing like a piece of art, crafting sentences with care and passion.
A Dialogic Perspective on Music Sampling
Sampling, since its emergence in popular music in the 1970s, presents somewhat of a grey area in terms of authorship. Although its ubiquity in present day music indicates wide spread acceptance of sampling as a compositional technique, the question of who is responsible for creating what is still very much open. This essay does not seek to propose any definitive answers to this issue, but only to offer an alternative perspective on the place of a sampled work with respect to others, as opposed to an isolated entity.
Analysis of music tends to explore it from one of two viewpoints: either from the ‘inside’, or the ‘outside’ (Korsyn, K., 2001). It can either analyse a piece as an isolated object, examining relationships between elements inside the piece, or as a result of a historical build up, considering its internal components as structures built from an axiomatic construction kit of classical precepts. An alternative perspective would be one rooted in Mikhail Bakhtin’s concept of dialogism (Irvine, M., n.d.). Though Bakhtin’s work deals principally with works of literature, the idea certainly applies to music. A simple way of putting it would be: “there are no texts, but only relationships between texts” (Bloom, H., 2011). The notion is that pieces can be viewed as if they weren’t pieces, but part of one larger entity, defined only by the connections between each other. Considering a piece like this allows an entirely different concept of what a sampled sound constitutes. This perspective can also be reached from a technical direction, as demonstrated by Curtis Roads in Microsound (Roads, C., 2001, p. 3). Although the main purpose of the book is to explore methods of conceptualising incredibly short sounds (microsound) and techniques for using them, to reach this point it was first necessary for him to define microsound, and as a result categorise all other sounds by length. To achieve this, he defines a number of time scales of music, including the ‘Supra’ time scale. Supra is defined as “a time scale beyond that of an individual composition and extending into months, years, decades and centuries.” Although Roads does not venture into abstract analytical territory, his supratemporal events can be taken as corresponding to relationships between lower level forms, such as individual pieces of music.
Sampling, in its current form, can constitute a plethora of compositional techniques, but the one thing they all have in common is the use of parts of a larger recorded sound to either create something new (to varying degrees), or present the sound in a different context. The term ‘sampling’ does not make a distinction between sounds recorded for the purpose of the composition, and sounds taken from other works, but in common use and in this essay, it refers primarily to the latter. One example would be taking a short segment of a piece, and looping it to provide the foundation for another, as in a commonly cited song in analysis of sampling in popular music, Vanilla Ice’s ‘Ice Ice Baby’, which samples the bassline of ‘Under Pressure’ by Queen and David Bowie. However, sampling is not limited to simply copying audio and using it elsewhere; manipulations of varying complexity are certainly possible. For example, a defining characteristic of Burial’s music is his use of detuned, timestretched vocals taken from popular music, such as ‘Archangel’, in which words from Ray J’s ‘One Wish’ are cut up and rearranged, ‘Ghost Hardware’ which features vocals from ‘Beautiful’ by Christina Aguilera, among other things. Indeed, much more advanced manipulations are possible using modern signal processing techniques, to the extent that the original sample may not even be recognisable by its original author. Apart from being simply processed alone, samples can be used as the seed of a sample- based synthesis engine, such as a granular synthesiser, to create something entirely different. It is important to note that samples can be taken from non-musical sources, such as film sound or spoken word recordings, and these can be treated in the same way as samples of music.
The existence of sampling is particularly useful for dialogic analysis of music as it arguably lends itself to this perspective. It is inherently viewable as a fairly pure connection; it involves the exact replication of a sound from another piece and its integration into something that may be completely different to its original context. Sampling allows for new types of connections between works that would not be possible without exact audio replication, though many forms of connection have been practice for a long time. Musical borrowing is an area that can be viewed as a set of dialogic relationships types, typically referring to classical music. For example, the final movement of Alban Berg’s “To the Memory of an Angel” borrows from a melody of J.S. Bach’s Chorale (“Es ist genug”).
Berg’s piece is a violin concerto, demonstrating that musical elements can easily be borrowed out of context to create something entirely different. By doing this, Berg was consciously referencing Bach, introducing a layer of interactivity with the listener on top of the rest of the music. While this particular link is fairly arbitrary, connections of a similar nature can be used to establish meaning, given prior knowledge of another piece. A good illustration of this is the ‘Tristan chord’, a chord used at the beginning of Richard Wagner’s ‘Tristan and Isolde’. The Tristan chord is often seen as an early seed of atonality (University of Texas, n.d.), and a great number of compositions have quoted it as an anchor. Using it almost implies that the composer is telling the listener that they intended to push the boundaries of tonality, and that they should accept it. It is a reminder of what has been done before, and gives automatic context to a piece. The Tristan chord has been used by Benjamin Britten (in ‘Albert Herring’), Claude Debussy (in ‘Pelléas et Mélisande’, and many others including Wagner himself in ‘Die Meistersinger’.
Given an illustrative traditional example like the Tristan chord, it is not difficult to imagine how sampling, employed in a certain way, could be used to exploit intertextual meanings lying hidden between pieces. Apart from simply referencing an element of a piece by just having it in another piece, a composer could apply some form of meaningful processing to it to evoke something that would not be accessible otherwise. A hypothetical example might be a composition that takes a vocal line from another piece and samples it. The vocal line could be laden with reverb indicative of a large, reflective room and detuned over time. Such processing could be said to create an effect that evokes a feeling of
dread. A possibility without sampling would be to reproduce the line in chorus with itself at a dissonant or non-musical interval. Performing such processing on a vocal line might shift the emphasis to words other than those originally emphasis in the source context, as changes in delivery of phrases can alter their linguistic emphasis in a way that is not uniform between words (Nolan, F., n.d.). By extension, this technique could be used to imply that the vocalist could be considered scary. Since the vocal line doesn’t necessarily have to be musical, it could also come from a political speech, for example, to imply something about the speaker, and in turn any pieces that may reference it in an alternate light.
Exploiting connections on the Supra time scale is something that is so common in sampling, that pieces that reference the act of sampling itself are often seen. To elaborate, sampling in early popular music that used it, such as hip hop, could often result in a ‘glitchy’ sound, as the techniques were not yet sophisticated enough to flow smoothly, and it was generally fairly obvious that something had been sampled. However, this became a stylistic feature and many artists still reference this in their music, even when the samples were recorded for the composition in question. This is heard in Bonobo’s ‘Kiara’, which features harshly chopped vocal parts that, even though they were recorded for the song, sound as if they were taken from something else. It could be taken as a statement that the artist was influenced by early sampling music, or at least connotes positivity for it. Of course, it would also have been possible for Bonobo to process the vocals in a humourous way, to imply that he was mocking the style. In fact, entire styles of music have been founded on this metareference, such as complextro, which makes use of glitchy processing on synthetic elements, making sounds that weren’t even recorded sound like they were sampled. IDM and glitch music also rely heavily on similar ideas.
This ubiquitious supratemporal self-awareness makes sampling a powerful illustrative tool for dialogism. Producers that employ sampling are inherently creating connections that reference other connections, meaning that monologically discussing such music is almost impossible. Add to this that different levels of understanding are unlocked by the listener upon hearing the original music that was sampled, and other music with similar processing, can be described well by another of Bakhtin’s terms: the ‘utterance’. An utterance is an instance of something, in Bakhtin’s original theory, words. As Bakhtin said: “There can be no such thing as an isolated utterance. It always presupposes utterances that precede and follow it. No one utterance can be the first or the last. Each is only a link in a chain, and none can be studied outside this chain.” Utterances are different every time that they are said, and depend on who the speaker and listener are, how they express it, where it is said, whether it has been said before and who said it then, and an infinite number of other variables. Samples, when viewed as an utterance, carry additional variables that are unique to sampled music. These include sonic characteristics only possible with modern audio processing, as well as variables pertaining to the act of sampling itself, not only in a technical way but as a reference to the technique and past uses of it. A sample as an utterance can also be used to highlight context without the interference of other aspects of the sample, as it can be perfectly reproduced without any modifications. In this case, the only thing that would be different is the context, which would make clear the connections between it and other work. Finding compositions where this is the case is rare, which may highlight an anxiety on composers’ parts to not directly copy another’s work. Harold Bloom also discusses this in his book, ‘The Anxiety of Influence’, albeit with emphasis on poetry rather than music. Bloom postures that artists are incredibly self-critical with respect to the relationship between their work and the work of previous poets, and strive for originality above all else. The main instances of artists sampling work with minimal technical modification usually involve non-musical sources. An example of this is Burial’s ‘Come Down to Us’, which samples a speech by transgender filmmaker Lana Wachowski with minimal modification. Using samples from non-musical sources like still allows a fairly transparent view of the supratemporal connections between works. In the Burial example, perhaps it says something about Burial himself. There is additional dialogic complexity introduced here by Burial as an artist; since few things about his personal life are known, strikingly meaningful utterances such as this become even more meaningful, like bright points on a dark backdrop. It could even be that Burial intended this expansion of the connotation strength spectrum.
Returning to the matter of anxiety, a cause of it could be over legal issues surrounding sampling and uncertainty over what constitutes fair use. Many artists have been sued for sampling others without permission, such as Danger Mouse’s ‘The Grey Album’ which sampled The Beatles, or The Notorious B.I.G.’s ‘Ready to Die’, which was pulled out of production because it sampled The Ohio Players’ ‘Singing in the Morning’. There exist entire companies dedicated to exacting legal action on users of sampling, such as Bridgeport Music, who own a large portion of George Clinton’s music (Keyes, A., 2012). From a dialogic perspective, ownership is irrelevant. All music can be seen as derivative. As Mikhail Bakhtin said, no one is “the first speaker, the one who disturbs the eternal silence of the universe.” There is a case to be made that ownership could be determined by the level of processing performed on a sample, and that might seem a logical way to do it. However, when sampling is viewed from a dialogic perspective, it is clear that elements other than sonic processing are in play. An artist can use a sample in a considerably different context to its source material, to generate a web of connections vast enough for that to be considered a creative act alone.
To conclude, considering any work dialogically allows for great insight into how it fits into the chain that brought it to existence, and sampling allows for additional materials for making links of the chain. It is difficult to truly come to an objective conclusion on authorship, and dialogism remains agnostic but perhaps mildly supports freedom of expression purely through virtue of not wanting to constrain itself. Or perhaps such constraints could be considered connections. Maybe the act of not sampling would imply something just as strongly as sampling, a void where a sample would be expected in similar music, refusing to conform to the style. That would be difficult to hear.
Korsyn, K. (2001). Beyond Privileged Contexts: Intertextuality, Influence and Dialogue. (Cook, N., Everist, M.), Rethinking Music (pp. 55-72). Oxford: Oxford University Press.
Bloom, H. (1973). The Anxiety of Influence. Oxford: Oxford University Press.
Bloom, H. (2011). The Anatomy of Influence: Literature as a Way of Life. Connecticut: Yale University Press.
Roads, C. (2001). Microsound. London: The MIT Press.
Kristeva, J. (1980). Desire in Language: A Semiotic Approach to Literature and Art. Columbia: Columbia University Press.
Irvine, M. (n.d.). Mikhail Bakhtin: Main Theories. Retrieved from: http://www9.georgetown.edu/faculty/irvinem/theory/bakhtin-maintheory.html.
University of Texas. (n.d.). Tristan and Isolde. Retrieved from: https://www.utexas.edu/courses/tristan/motive1-3.php.
Nolan, F. (n.d.). Intonation. Retrieved from: http://www.ling.cam.ac.uk/francis/FN_inton_prepub.pdf.
Keyes, A. (2012). George Clinton Fight For His Right To Funk. Retrieved from: http://www.npr.org/blogs/therecord/2012/06/06/154451399/george-clinton-fights-for-his- right-to-funk.