Collaborative Songwriting – academic paper
This is an academic paper on the subject of collaborative songwriting in the studio. It was presented at the 6th Art of Record Production conference in Dec 2010 and appears in the Journal of the Art of Record Production Conference Proceedings – ISSN 1754-9892. Please feel free to download/cite it as you think fit. The correct citation is;
Bennett, J., 2011. Collaborative songwriting – the ontology of negotiated creativity in popular music studio practice. In Journal of the Art of Record Production 2010. Leeds, UK: Art of Record Production.
In keeping with my view that academics should make their research as freely available as possible, you can download the whole paper here.
Collaborative songwriting – the ontology of negotiated creativity in popular music studio practice
Bath Spa University
The relationship between songwriting practice and song product is an under-explored one in popular musicology, still less so in a studio-based environment. Our research sources are accordingly limited, drawing mainly on first-hand retrospective interviews with artist-songwriters, who may have an incentive for self-mythologising, or at least romanticising their songwriting methods to preserve fan perceptions of authenticity. There are no available real-time observations of the collaborative processes involved in creating popular song, despite the huge economic and artistic successes of songwriting partnerships throughout the history of our field. Sloboda (1985) identifies the reluctance displayed by composers of any sort to participate in detailed analyses of their processes; these difficulties are exacerbated further by some songwriters’ apparently deliberate mystification of their craft. Attempts to analyse processes of musical composition have generally focused on single-composer models (Nash 1955); even studies relating to collaboration remain concerned with instrumental art music (Hayden & Windsor 2007) or educationally-based observation subjects (Burnard & Younker 2002).
This paper builds on the single-songwriter research of McIntyre (2009) and the theoretical definitions of creativity provided by Csikszentmihalyi (1996). It explores, through analysis of ‘hits’ and examples of emerging practitioner-based research, the inferences that can be made by comparing historical and current songwriting practice with the finished product, and attempts to identify commonly-used collaborative models, including a discussion of the effect of the presence (or absence) of studio technologies as mediator of the songwriting process.
Popular song is one of a handful of unsubsidised populist art forms (other examples being mainstream cinema or video games) that could be described as truly market-driven. The market validates works (e.g. download charts, gig attendances or YouTube clicks) and provides the economic circumstances for creators to make new work (through royalties to the songwriter). Many other art forms – opera, contemporary dance, sculpture, fine art, poetry, art music, and some theatre – are not economically self-sustaining without some form of subsidy in addition to the consumer’s own contribution. This difference in economic context provides a different creative climate – perhaps a paradox – for the songwriter, who is trying to create an original work within a highly evolved, market-driven, and tightly constrained creative palette.
Thus, some Darwinism is at play in the metrics of gig attendance, CD sales charts, airplay and downloads. Unsuccessful songs die due to lack of interest, and successful songs survive to be heard by many listeners. The evolutionary metaphor could be taken further – elements of originality in a successful song (let’s call this originality a ‘genetic mutation’) may inspire other songwriters to incorporate these ideas into future songs. The celebrated psychologist Mihalyi Csikszentmihalyi applies this Darwinist model to all creativity, describing the palette of accepted creative ideas as the ‘domain’ in his Systems Model of Creativity (1988). Domains are validated by a ‘field’; his catch-all definition of creativity describes this as a ‘field of experts’, although in the case of popular song the experts in question are consumers, albeit mediated by the mechanics of the music industry’s pre-selecting gatekeepers (A&R, record companies, music publishers, radio playlisting etc). The third element of Csikszentmihalyi’s system is the individual; if a creator invents a field-validated object that then survives to join the domain, the individual has been truly (capital C) ‘Creative’. If we apply the Csikszentmihalyi model literally to the world of popular song then perhaps only enduring classics are truly Creative; I suggest that the definition of (lower-case c) creativity in songs has a lower threshold. There are many songs that are ‘original’ works (in the legal sense) that may become economically successful, but this does not necessarily mean that they will become influential in the domain of songwriting.
2 Popular song – definition by constraint
The economic mechanisms that drive audience approval of songs have another important effect – they shape the art form itself. This contention is framed by the assumption that all art forms (that can be categorised) are at least part-defined by their constraints – a haiku being a prosaic but usefully simple illustrative example. Popular songs have, through audience-driven ‘natural selection’, evolved many characteristics in common with each other that, I suggest, define the form, or at least the popular mainstream of which less (literally) popular niche genre-songwriting activities form tributaries. The table below (Fig 1) demonstrates a selection of these, compiled through my analysis and observation of the Anglo-American singles and albums charts 1954-2010. This is not to say that all songs will exhibit these characteristics; rather, a majority of them will appear in almost all successful songs, and some mainstream classics will have most or all of them. For example, at the time of writing (October 2010) the current UK number 1 download is Bruno Mars’ Just The Way You Are – a song that exhibits 100% of the characteristics, as do the majority of songs in the current top 10. A comparison to the equivalent top 10 from any decade since 1960 gives much the same results, allowing us to speculate that some of the constraints that define song form may be constants, at least in mainstream hits of the last 50 years.
Fig. 1 Common characteristics of mainstream hits
- First-person sympathetic protagonist/s, portrayed implicitly by the singer
- Repeating titular choruses (where the song is in chorus form), usually containing the melodic pitch peak of the song, which summarise the overall meaning of the lyric
- Rhyme – usually at the end of lyric phrases
- One, two or three human characters (or a collective ‘we’)
- Feature an instrumental introduction of less than 20 seconds
- Include the title in the lyric
- Sung between a two-octave range from bottom C to top C (C2 to C4), focusing heavily on the single octave A2 to A3.
- Thematic lyric content relating to (usually romantic) human relationships
- Use underlying 4, 8 and 16 bar phrases, with occasional additions or subtractions
- Based on verse/chorus form or AABA form
- 4/4 time
- Maintaining one diatonic or modal key
- Between 2 and 4 minutes in length
As McIntyre (2001) implies, form constraints are understood by experienced songwriters; they are the landscape in which song originality thrives. Indeed, all musical forms could be said to be defined by constraints, and these may actually increase creative opportunity, as Stravinsky notes;
“…my freedom will be so much the greater and more meaningful the more narrowly I limit my field of action and the more I surround myself with obstacles. Whatever diminishes constraint diminishes strength. The more constraints one imposes, the more one frees one’s self of the chains that shackle the spirit.” (Stravinsky 1942)
When two or more songwriters collaborate, they will share a desire for their song to be heard by others; this is frequently economically-driven, but also born of a creative and artistic goal – to make an object that communicates emotionally. In the case of non-performing songwriters an additional factor comes into play – the song has to be suitable for a particular (or hypothetical) performer, who may or may not be part of the songwriting team. Mainstream songwriters who work with a variety of other collaborators may have an implicit understanding of the norms of songwriting, and write songs that adhere to most of these norms, breaking occasional constraints according to taste. Jez Ashurst, a songwriter who has written for Boyzone and other pop artists, describes his relationship with form thus;
“You want the chorus to come in between 40 seconds and a minute and you’ll want the song to be [ending] within three to four minutes […]. If you get to the end of the second chorus, and you’re with another writer, they’re not going to say [to each other] “what happens now?”. You’ll know that you’re going to do something for about 16 bars and go back to a final chorus. It’s almost a given. [So as a songwriter] you know the shape of the box, and you’re happy to be in that box.” (Ashurst 2010a)
Moore (2009) introduces musical ‘invariants’ that define particular songs, and suggests that the listener’s interpretive meaning is created within these in the context of each song. It is hard to dispute that individual songs exist within a set a defined musical constraints, be they tempo, key, lyric theme, plot, form or characters. I further contend that we can apply Moore’s ‘invariants’ to popular song form itself if we use the word less specifically i.e. allow for a selective approach, where songwriters create meaning by choosing particular invariants from song to song from a culturally understood but necessarily limited menu such as the one I have provided (Fig 1).
“In music, it is certainly possible to identify those constants which remain necessary to the performance of a particular song, and which remain present from one performance to another. Indeed, I have insisted elsewhere on a distinction between three categories – song, track and performance – and, at one extreme, invariants seem to be those very characteristics which define a particular song. At the opposite end, the binary metre of both ‘Good Vibrations’ and ‘See My Friend’, or their tonal centres, operate as invariants against which the constant change of individual durations, or of individual pitches, creates meaning…” (Moore 2009)
3 Defining process
My research into the creative processes undertaken by collaborative songwriters centres on two questions; what processes can contribute to a successful song, and why should a songwriter collaborate with others (when there is an economic disincentive to do so due to sharing of royalties).
The song is the script from which the rest of the popular music production chain reads; it leads to the recording session, the artist’s vocal performance, the promotion and distribution of the final mixed audio artefact, and thereafter implicitly to tertiary activities – gigs, airplay and merchandising. Without a ‘good song’ at the heart of these processes, the music industry cannot function. That makes songs into inherently valuable Intellectual Property – publishing copyrights are negotiated, traded, and jealously guarded through necessarily pedantic plagiarism lawsuits. And yet, compared to other cultural objects, songs in their pre-recorded form are ostensibly quick to manufacture, musically simple, and literarily uncomplicated. This tension between high (economic and cultural) value and apparent ease of manufacture leads me to my first question – why, if songs are so important, is the process of their creation so mysterious and undocumented?
Part of the answer is that the mysteriousness itself is a cultural asset i.e. it is desirable for some songwriters, particularly singer-songwriters, to shroud their craft in romance and mystery. The majority of interviews with songwriters obviously feature those who are also artists, who will have an artistic persona to sell, and therefore a motive for concealing more mundane, contrived or even random aspects of the composition process that may be perceived by fans as unromantic. Many contemporary artists, even (current UK) singer-songwriters like Katie Melua, James Morrison, Lily Allen, James Blunt and Newton Faulkner, actually use backroom co-writers, but are incentivised to obfuscate their collaborative processes because of the need to sell the authenticity of the song – and therefore their own credibility as ‘songwriters’.
The other challenge is that songwriting is not easy to document in practice. The first difficulty is finding songwriters who will agree to be observed, followed by the need to construct an observational environment that minimises the risk of damaging the process due to the observer-expectancy effect. Even if these hurdles are overcome, the various mechanisms of observation – audio recording, text transcription, immediately-retrospective interview, even live video – may not fully capture all of the creative forces at play in a co-writing session. Finally, even when a detailed observation of a songwriting session is complete, the statistical likelihood is that the song will not be commercially successful – making the documentation of Csikszentmihalyi’s ‘large-C’ Creativity ever more elusive i.e. we cannot know the cultural value of the emerging creative object until long after our observation and analysis is completed.
However, there is an attendant advantage to studying collaborative as opposed to solo songwriting: all creative ideas must manifest themselves in order to be communicated to the other writer. In becoming manifest, albeit after some level of internal veto by their creator, these creative ideas become observable. Further to this, I am acting as co-researcher by collaborating in many of the songwriting sessions myself, enabling me to experience the decision-making, negotiation and veto processes first-hand.
This process of editorial veto is essential to the collaborative songwriting process, and may be one of the reasons that it is such an historically successful creative model. Songwriters often describe their writing partner as a ready-made audience, an extra pair of ears or similar phrase (Carter 1990, p.4). The approval of a creative idea by the co-writer potentially doubles its chances of being a ‘good’ idea and risk-manages the subjectivity of the initial creator. This instant-audience effect, combined with the fact that more ideas can presumably be generated in a collaborative environment, may increase a song’s chances of success compared to a solo-written work.
At the time of writing (October 2010), the UK pop singles market is dominated by collaboratively written songs; in this environment at least, it is the preferred contemporary industrial model for creating successful hits. Four of the current UK top 5 singles (iTunes 2010b) are written by collaborative teams. This is admittedly less true of the ‘album market’ where bands and singer-songwriters are more likely to be found, but as previously stated, even these artists sometimes have a surprising level of input from professional co-writers – and of course there are many ‘authentic’ legendary bands (U2, REM, Led Zeppelin, The Beatles) where the songs were co-written – or at least, were credited as such.
There have been many thousands of collaborative partnerships in the history of Anglo-American popular songwriting, but my ongoing work triangulating interviews and observations reveals a surprisingly small number of collaborative models, which despite some degree of overlap I attempt to summarise thus;
Nashville. Acoustic guitars/piano and minimal technology – a ‘pen and paper’ approach typically featuring two writers, who usually do not have demarcated roles.
Factory. A geographical location with staff songwriters; notable examples include Tin Pan Alley (late 19th/early 20th Century), The Hit Factory (1980s), The Brill Building (1950s/60s) and Xenomania (2000s). Like the Nashville model, it is defined in part by a regimented timeframe – songwriters ‘come to work’ in the morning. ‘Factories’ may use parts of the other models, and are currently frequently studio-based (Higgins 2009).
Svengali. The artist is one co-writer, although their input may vary from a small contribution such as a title through to a substantial one such as a complete lyric. Typically the other co-writers are more experienced than the artist, and may have collaborated with a large number of others. A recent UK example is James Blunt’s breakthrough hit ‘You’re Beautiful’, which was co-written (Hewson 2009) with two professional songwriters with a prior track record of hits.
Demarcation. A lyricist provides a finished lyric for word-setting by a composer (e.g. John/Taupin or Difford/Tilbrook), or the composer provides music for a lyricist to write to (e.g. Mercer/Mancini). This model is unusual because the parties need not meet in order to co-write – in this sense it is only arguably collaborative because it does not usually provide veto or negotiation. The line of demarcation need not be split between music and lyric (although this is historically the most common approach) and could, for example, be split between melody and harmony, or in studio teams, tonal material and drum programming.
Jamming. A band creates live ideas in the rehearsal room, forming the song from individual contributions to the arrangement and some degree of veto (e.g. U2). Band members may bring stimuli to the session (titles, riffs etc).
Top-line writing. A completed backing track is supplied by a ‘producer’ to a top-line writer who will supply melody and lyric. The backing track acts as harmonic/tempo template but more crucially as inspiration for genre-apposite creative decisions, such as singability of a line.
Asynchronicity. The co-writers work separately and iteratively, but do not necessarily define clear or exclusive creative roles. An example would be if two songwriter-producers worked separately on a multi-track audio file, passing it backwards and forwards (typically online) and making iterative changes in one or more cycles. The demarcation model is usually implemented asynchronously, but asynchronous writing need not be demarcated (by activity or creative contribution).
So perhaps we are beginning to form a picture of why and how songwriters collaborate, and we may have an idea – through decades of successful song form evolution – of the shape of the object they intend to create. But what do they do? What happens in that private, undocumented environment when two or more individuals – who may not even know each other – sit down together to create a new song from nothing?
The first answer is that the song is rarely created literally ‘from nothing’. Stimulus material, however small, usually provides a starting point for the creative process (Carter 1990, pp.44-51). Nashville co-writers typically turn up for work with a number of titles. Band-written songs may begin with a riff on a particular instrument. Contemporary pop writing usually includes an element of technology: many songwriters, individual or collaborative, use drum machines or computer equivalents to provide a temporary backing track over which to try ideas. Professional non-performing songwriters who are co-writing with a less experienced songwriter-artist – a very common model in contemporary UK pop songwriting – will often bring some pre-prepared ideas to the session (Ashurst 2010a; 2010b).
I contend that six (non-linear and interacting) processes are at play in a co-writing environment – stimulus, approval, adaptation, negotiation, veto and consensus. One writer will provide stimulus material and the other writer will approve, adapt or veto the idea (approval can obviously lead to consensus – I include both because there may be situations with more than two co-writers where one individual approves an idea but another provides veto or adaptation). If an idea is vetoed in its entirety the provider of the stimulus will either accept this, or enter negotiation to defend or further adapt it. Consensus permits an idea to survive and – temporarily or permanently – take its place in the song (collaborative songwriters frequently report agreement that “we’ll fix that bit later” – for example, in the use of a dummy lyric that will later be replaced). This theoretical model can be applied to any collaborative songwriting process or practice (and perhaps to other collaborative artforms), and may include complex interactive behaviours that are difficult to observe. For example, the ‘testing’ of a stimulus idea may include an element of improvisation between the writers. This is not true improvisation as defined and investigated by Sawyer (2003) because it is not fully public; rather, it is a form of adaptation and ‘play’ that may lead to approval, veto, consensus or further adaptation.
5 Process, product and the studio
There is an apparent relationship between the mechanics of process and the characteristics of the finished product, and this can be dramatically affected by assistive technologies. To choose a prosaic and unsubtle example, a large amount of early 1990s Dance and House tracks (let’s call them ‘songs’) have a tempo of exactly 120 beats per minute. This is because the software template of one of the most influential pieces of software – Cubase – defaulted to this tempo in 1989 when it was launched, increasing the likelihood of composers using it in their work. But less obviously, particular working methods create a different creative environment in which certain musical decisions – and therefore outcomes – become more likely. Ashurst (2010a) chooses instruments according to the desired genre – piano for ballads, acoustic guitar for singer-songwriters, electric guitar for rock.
And this is where the studio or music computer plays a significant part in the creative process – it becomes arguably the ‘invisible writer’ in that it generates or enables stimulus material that would not be part of the creative process in a non-studio environment. Drum loops are an obvious example; in my own teaching of songwriting I have observed that collaborating students usually write at higher tempi (120BPM and beyond) if they use drum loops than if they use guitar or piano; I infer this is due to the physical ergonomics of strumming up and down-strokes on a guitar, and may be related to resting heart rate being less than 120. This in turn affects note scansions – faster tempos are more likely to syncopate to a quaver than to a semiquaver because the lyrics are more difficult to sing over complicated syncopations at higher tempi. Speed of scansion will also affect vowel behaviour in the composition of lyrics for sung phrases – multiple complex consonants (plosives and sibilants) are more difficult to sing at speed; thus, using a drum loop while writing could have an effect on the vowel/consonant balance in a lyric.
Like any musical tool, studios – and particularly computer-based sequencers – encourage particular ergonomic habits in songwriters, creating musical outcomes that find their way into the song and eventually the songwriting genre. ‘Loop’ songs (i.e. songs where the entire harmony consists of three or four chords played in 2, 4 or 8-bar loop) are noticeably more common in contemporary UK pop than they were in the equivalent 1960s charts, and they are in turn more common in artists working in computer-based genres (R&B, Hip-Hop) than in band-based artists. To return to our current (October 2010) top 5 UK singles, four are based almost entirely on four-chord loops; one is based on a three-chord loop. Computer sequencer software favours loop-based songwriting because looping of musical material is less time-consuming for the operator than writing new material, and also because the ‘cycle’ mode encourages users to hear the same short excerpt (typically four bars) over and over while making edits. Interestingly, the specialist rock chart for the same week (iTunes 2010a) shows similar looping characteristics in three of the top five songs, the two exceptions being reissues of songs composed before music software was in common use by songwriters (Jonah Lewie’s ‘You Will Always Find Me In The Kitchen at Parties’ and Journey’s ‘Don’t Stop Believing’, both recently re-popularised by an appearance in a TV advertisement and TV show respectively). In the self-referential and self-influencing world of popular music, we can infer perhaps that this observed increase in the number of loop-based songs has ‘jumped species’ from computer-based to band-based genres, even though the ergonomic driver did not come purely from band-based songwriting. Thus, even though an individual four-chord loop song may not itself become part of Czikszentmihalyi’s ‘domain’, the practice of looping certainly has done so.
So if we know that process affects product, we can to some extent reverse-engineer the product and make educated guesses at the way particular songs might have been composed. In turn, songwriters, empowered by experience and listening, with an awareness of the relationship between process and product, can to some extent influence the nature of the song by using processes that are more likely to achieve a desired outcome; thus, process selection becomes a creatively meaningful decision in itself.
6 Separating ‘song’ from ‘track’
There was a time in Anglo-American popular music’s history (the first half of the 20th century) when the song and recording were entirely separate objects. Songwriters’ activities culminated in the publishing of sheet music, which was then available to professional interpreters or to the general public as a retail product. Music publishers (a term that still persists today despite the relative unimportance of score publishing) protected and administrated royalties, but even in the early 1950s the sheet music market was more significant for songwriters than the recorded one – Lindsay (1955) speculates as to whether the gramophone market will ever become more important to songwriters than sheet music, and idea which seems naive today after more than five decades of ‘hit singles’.
In these early years, the songwriter usually did not concern themselves with timbral aspects of the arrangement, still less creative input to the recording process. Thus, the musical artefacts of value in the song were forced to focus exclusively on musically desirable characteristics that could be notated i.e. melody, harmony and lyric. To take harmony as an example, it is interesting to hear the increasing harmonic sophistication (via extended chords and ‘substitutions’) in songs between early ragtime and the evolution of the ‘jazz standard’. Like any musical trend, this ran its course, and popular songs found a new harmonic simplicity from the late 1950s with the advent of rock ‘n’ roll; harmonic sophistication then became a minority interest, genre-compartmentalised in jazz.
Moore (2009) has alluded to the difficult necessity of separating ‘song’ from ‘track’ – the dilemma also preoccupies Tagg (2009), who despite being critical of a perceived graphocentric (score-based) approach in some popular music analysis, does not propose a workable alternative. McIntyre (2001) elaborates on the problem;
“The very term ‘song’, once investigated, becomes problematic… Copyright law won’t tell us precisely [and] there are indications that the audiences for songs also may have a wide-ranging concept of what may constitute a song. Importantly, musicians who deal with songs every day have an assumed but increasingly elastic definition.” (McIntyre 2001)
The 21st-century collaborative songwriting team, working in a contemporary studio using a computer-based workstation, has become increasingly empowered to blur the lines between songwriter, arranger, performer and producer. Studio tools themselves may act as the stimulus for creative processes; a recent example can be found in Rihanna’s 2007 hit Umbrella, where a factory-supplied audio sample in Apple’s free/bundled Garageband software became one of the defining sonic artefacts, not only as a (possible) starting point for the songwriting process but also appearing in the finished track (Webb 2007).
From the mid-1980s to the present it is easy to chart a democratisation of studio-based songwriting due to falling equipment prices and increasingly powerful technologies; gone are the days when a studio’s worth was partly defined by its multi-timbrality (i.e. number of ‘tracks’). The Beatles’ 1960s sonic innovations, being based in expensive (and staffed) studio facilities, were possible because of the wealth generated by early hits that required a comparatively smaller outlay, so funding more experimental and expensive studio time for this artist was, by the mid-1960s, a logical investment. In the 21st century, sonic innovation, or at least access to massively powerful production tools, is now available to all at negligible cost. This ubiquity means that when songwriting and production overlap, as they do in so many sub-genres of popular music, studio-based production skills will increasingly affect the outcome of the collaborative songwriting process.
Perhaps this democratised production climate has forced songwriters to engage with the fundamental skills that originally drove the 20th century music industry – the ability to write a ‘great song’. In this context, many songwriters have taken the decision that the benefits of collaboration outweigh the loss of income – partly because they believe that they will write a better song in this environment. Despite many changes in the means of production of popular music in the last 60 years, the practice of distributed and shared creativity in songwriting continues to thrive as one of its defining forces.
Ashurst, J., 2010a. Jez Ashurst interview – on collaborative songwriting (April 2010). Interviewed by Joe Bennett.
Ashurst, J., 2010b. When The Well Dries Up. Myspace blog – Jez Ashurst. Available at: http://blogs.myspace.com/index.cfm?fuseaction=blog.view&friendId=202597070&blogId=531729156 [Accessed May 8, 2010].
Bell, L., 2011. International Skype Songwriting! | Lisa Bell Music Singer/Songwriter. Lisa Bell – songwriting blog. Available at: http://lisabellmusic.com/2011/01/international-skype-songwriting/ [Accessed May 4, 2011].
Bennett, J., 2009. Crows, Rooks and Ravens – the songs Â« Joe Bennett. Joe Bennett website. Available at: http://joebennett.wordpress.com/2009/03/09/crows-rooks-and-ravens-the-songs/ [Accessed May 4, 2011].
Bennett, J., 2011. How long, how long must we sing this song? Â« Joe Bennett. Joe Bennett music blog. Available at: http://joebennett.wordpress.com/2011/05/03/how-long-how-long-must-we-sing-this-song/ [Accessed May 3, 2011].
Boden, M., 2004. The creative mind : myths and mechanisms 2nd ed., London ;New York: Routledge.
Burnard, P. & Younker, B.A., 2002. Mapping Pathways: fostering creativity in composition. Music Education Research, 4(2), pp.245-261.
Available at: http://search.ebscohost.com/login.aspx?direct=true&db=ehh&AN=7401630&site=ehost-live.
Carter, W., 1990. Writing together : the songwriter’s guide to collaboration, London: Omnibus.
Clydesdale, G., 2006. Creativity and Competition: The Beatles. Creativity Research Journal, 18(2), pp.129-139. Available at: http://www.informaworld.com/openurl?genre=article&doi=10.1207/s15326934crj1802_1&magic=crossref||D404A21C5BB053405B1A640AFFD44AE3.
Csikszentmihalyi, M., 1996. Creativity: Flow and the Psychology of Discovery and Invention, New York: HarperCollins.
Csikszentmihalyi, M., 1988. Society, culture, and person: a systems view of creativity. In R. Sternberg, ed. The Nature of creativity : contemporary psychological perspectives. Cambridge ;;New York: Cambridge University Press, pp. 325-339.
Gibson, J., 1986. The ecological approach to visual perception, Hillsdale N.J.: Lawrence Erlbaum Associates.
Hawkins, S., 2003. Feel the beat come down: house music as rhetoric. In Analysing Popular Music. Cambridge University Press. Available at: http://www.tagg.org/others/Hawkins/HawkHouse.html [Accessed November 1, 2010].
Hayden, S. & Windsor, L., 2007. Collaboration and the composer: case studies from the end of the 20th century. Tempo, 61(240), p.28. Available at: http://www.journals.cambridge.org/abstract_S0040298207000113.
Hewson, J., 2009. Songbook: James Blunt. Songbook: James Blunt. Available at: http://www.skyarts.co.uk/music/article/songbook [Accessed February 2, 2010].
Higgins, B., 2009. Brian Higgins interview – The Telegraph (August 2009). Available at: http://xenomania.freehostia.com/press/brian_telegraph_aug09.html [Accessed January 31, 2010].
iTunes, 2010a. Apple – iTunes – iTunes Store – Charts – Top 10 Rock Songs. Available at: http://www.apple.com/euro/itunes/charts/top10rocksongs.html [Accessed October 23, 2010].
iTunes, 2010b. Apple – iTunes – iTunes Store – Charts – Top 10 Songs. Available at: http://www.apple.com/euro/itunes/charts/top10songs.html [Accessed October 23, 2010].
Lindsay, M., 1955. Songwriting Rev. impression., London: Teach Yourself Books.
Mcintyre, P., 2009. “I”m Looking Through You’: An Historical Case Study of Systemic Creativity in the Partnership of John Lennon and Paul McCartney. In Collaborations: Creative Partnerships in Music. The Performance and Social Aesthetics Research Unit (PASA), Monash Conference Centre, Monash University, Melbourne, Australia.
McIntyre, P., 2001. The Domain of Songwriters: Towards defining the term “Song.” Perfect Beat: The Pacific Journal of Research into Contemporary Music and Popular Culture, 5(3), pp.100-111.
Moore, A.F., 2009. Interpretation: So What? In D. B. Scott, ed. The Ashgate research companion to popular musicology. Ashgate.
Nash, D., 1955. Challenge and Response in the American Composer’s Career. The Journal of Aesthetics and Art Criticism, 14(1), pp.116-122. Available at: http://www.jstor.org/stable/426646.
Pettijohn II, T.F. & Ahmed, S.F., 2010. Songwriting Loafing or Creative Collaboration?: A Comparison of Individual and Team Written Billboard Hits in the USA. Journal of Articles in Support of the Null Hypothesis, 7(1), p.2.
Sawyer, R., 2003. Group creativity : music, theater, collaboration, Mahwah N.J.: L. Erlbaum Associates.
Sloboda, J., 1985. The musical mind : the cognitive psychology of music, Oxford [Oxfordshire] ;New York: Clarendon Press ;;Oxford University Press.
Stravinsky, I., 1942. Poetics of music in the form of six lessons, Cambridge Mass.: Harvard University Press.
Tagg, P., 2009. Everyday Tonality, New York & Montréal: Mass Media Scholars’ Press (via tagg.org).
Webb, A., 2007. Is GarageBand top of the pops? | Technology | The Guardian. Available at: http://www.guardian.co.uk/technology/2007/oct/18/news.apple [Accessed November 1, 2010].
Zollo, P., 1997. Songwriters on songwriting, Da Capo Press.
 Boden (2004) defines this manufacture of historically significant works as ‘H-creativity’ and contrasts it with ‘P-creativity’, where a creative idea is psychologically new to an individual but not necessarily new to world history. Thus, all H-creative ideas are P-creative, but not all P-creative ideas are H-creative.
 Song duration in successful hits is one of the easiest constraints of form to illustrate with hard data (Bennett 2011); chart/sales analysis of these data can also provide evidence of ‘evolutionary’ trends over time.
 Moore cites JJ Gibson’s model of ecological perception (1986), and for the purposes of this discussion my ‘constraints’ could be described as a Gibsonian ‘environment’.
 Blunt can be heard ‘backgrounding’ the contributions of his professional co-writers in interviews (Hewson 2009).
 One recent application of the co-researcher role in the observation of collaborative composition can be found in Hayden & Windsor (2007).
 Pettijohn II and Ahmed (2010) found in a longitudinal study of the Billboard charts 1955-2009 that collaborative teams were responsible for as many number 1 hits as individual songwriters.
 Paulus’ findings do not necessarily contradict this assertion; the generation of twice as many ideas may simply increase productivity. Economic Darwinism is still a deciding factor in the eventual success of the end product, regardless of the circumstances of its creation.
 This distinction between genuine and ‘name-only’ co-writing is particularly important in the case of Lennon/McCartney, where very few songs were actually co-written in the sense of a two-part creative collaboration; rather, the co-writer’s role was primarily one of veto. McIntyre (2009) discusses this in more depth, as does Clydesdale (2006).
 See also Zollo (1997, pp.656-658)
 A remarkably precise illustration of Tagg’s melody/accompaniment dualism (Tagg 2009, p.268).
 Internet technologies have facilitated this model greatly since the 1990s, and there is some emerging evidence (Bell 2011; Bennett 2009) that synchronous/real-time online technologies (e.g. Skype) may be playing an increasing part in this model and others, allowing more opportunities for online implementation of the ‘traditional’ models.
 Hawkins (2003) reasonably uses the more contemporaneous word ‘track’ to describe House music, although during recent years ‘tune’ is becoming increasingly fashionable term for some forms of mainly instrumental dance music in the UK.
 A function in music recording software that enables a particular section to be repeated ad infinitum by the operator.
 This approximate forty-year period is arguably a close ‘evolutionary’ parallel to the ever-increasing tonal sophistication that developed over 500 years in European art music, from the Renaissance to Serialism.