Kinds, Covers, and Kinds of Covers

© 2022 P.D. Magnus, CC BY 4.0

At the end of the last chapter, I refused to give a definition of ‘cover’ but promised that there were other important distinctions to make. In this chapter, I distinguish songs, performances, and tracks— arguing that all three are objects worthy of appreciative attention. I then make a distinction between two kinds of covers: mimic covers and rendition covers.1

Musical works

We can easily distinguish between a song itself and performances of that song. For example, the song ‘Happy Birthday’ can be sung by different people on the occasion of different birthdays. When it is sung on a particular occasion, there is one performance of it. There may be many different performances simultaneously, in different places around the world, to celebrate different people’s birthdays. All these different performances are performances of one and the same song.

Some recordings are just intended to provide access, as much as possible, to an original performance. For jazz records, the entire ensemble typically plays together. The recording is taken from just one performance rather than multiple performances mixed together. Often, the best take is chosen for release. In some cases, multiple takes are released— as on the Miles Davis album Bag’s Groove, which begins with two takes of the title tune. Jazz listeners want to hear the performance, and the recording is expected to transmit that as much as possible.

In contrast, recordings in rock and pop music typically do not correspond to any single performance. Different musicians record their parts separately, rather than together. For each part, the final track includes material from multiple separate performances. Sometimes these are performances that could not have occurred together, as when Stevie Wonder recorded four different instrumental parts and vocals for ‘Superstition.’ Not only are the best bits selected and assembled, they may also be transformed by various effects. An egregious example of this is the heavy use of Auto-Tune, software originally introduced in the late 1990s, which makes sounds match a prescribed pitch but can also make them sound artificial and robotic. Furthermore, samples and sounds can be introduced in production that do not correspond to sounds which were ever made by performers.

Theodore Gracyk notes that a track which is the result of editing and post-production in this way ‘is accessible only by playing, on machines, end-products derived from a montage of partial performances’ (1996: 34). For rock music, the recorded track is the way that most people encounter the song. It is what they hear on the radio and (depending on the time period) on records, tapes, CDs, or streaming internet audio. As Gracyk puts it, tracks ‘are the standard end-products and signifiers in rock music’ (1996: 36).

What he has in mind is the master track or copies of the master. Talking about recording and production, the word ‘track’ has another use. Different parts or sets of sounds are recorded to different tracks. When recording is done to tape, the tracks are literally paths on the tape itself. With computers, they are separate sound files. These are mixed together for the final version. When the contrast is not at issue, the final version is simply called a track— as when discussing the tracks on a published album. I will not have much cause to talk about separate recorded parts, so I will use ‘track’ to mean the finished product.

This makes for three different kinds of things:

  1. Lyrics and musical structure provide the identity conditions for the song. This allows for different interpretations and arrangements, so that performances and recordings of the same song can sound rather different.

  2. A performance is an event which happens only once. Although we can listen to a recording of it later, the same performance cannot happen again. A repeat performance, even if it sounds the same, is a separate instance of the same song.

  3. A track is a specification of sounds. It can exist as multiple copies in different media, and it can be played back repeatedly.

Note that these are explications in the sense discussed in the previous chapter. Making the distinctions in this way allows me to precisely discuss particular issues, but it does not perfectly match ordinary usage. It is common to refer to a cover as a ‘cover song’ even though a cover is a performance or recording rather than a song in the sense distinguished above.

Contrast the current Apple Music Style Guide, which defines ‘Song’ as ‘An audio recording’ and ‘Track’ as ‘A song or music video’ (2021). The Spotify Metadata Style Guide is similar. These definitions do not fit common usage, either. Instead, they are specifications for the purpose of content metadata. Like mine, these definitions are explications. It is just that their purposes are different than mine. Apple and Spotify want to maintain enormous digital storehouses of media files, whereas I want to understand covers.

Even experts will often not mark the distinction between what I am calling songs and tracks. Producer and educator Rick Beato comments in a YouTube video discussing a recent hit, ‘That’s a very well produced song. It’s got a great melody. Interesting twist in the chord change. Whoever mixed it, the low end is massive’ (2021). The melody and chord changes are features of the songwriting, but the mix and the low end are features of the recording. Similarly, a list of ‘500 Greatest Songs of All Time’ in Rolling Stone slips from high-level considerations of lyrics and melody to low-level considerations of timbre and production (2021). The point is not that Beato or the writers at Rolling Stone would not recognize the distinction between songs and tracks, but rather that they do not always mark the distinction in reacting to music.

Admittedly, none of these things would show up on a list of what fundamentally exists. Neither songs, performances, nor tracks are mentioned in physicists’ Grand Unified Theory— nor do they appear on philosophers’ lists of fundamental entities or categories. Nevertheless, songs, performances, and tracks all exist in an ordinary sense of exist.

The boundaries of each category will not be perfectly precise, however, and the process of writing a song, performing it, and constructing a track can merge together. When a musician noodles around on an instrument, toying with a riff, there is no sharp divide between the playing and the songwriting. If they record themselves doing that, then there may be no sharp line when songwriting ends and constructing the track begins.

Albin Zak discusses the example of the Beatles’ ‘Strawberry Fields Forever.’ It was originally introduced to the band with only guitar and vocals. A version was then arranged and recorded with the whole band. Setting that version aside, producer George Martin wrote scores for cello and trumpet parts, which were recorded along with additional percussion parts. The tape of the percussion instruments was reversed, making a kind of sucking sound. The resulting version is ‘a layering of nonsynchronous performances through overdubbing, and, in the backwards percussion, timbral characters produced by the recording medium itself.’ Ultimately, the decision was made to use both versions. By slowing down one and speeding up the other, the pitch and tempos were shifted ‘to make a credible, if not altogether unobtrusive, edit’ (2001: 36). The resulting track is the band version for about a minute and then shifts to the more complicated second version. In cases like this the process of composition is interwoven with the process of recording.


Even recognizing that all three things exist, it is natural to wonder if they are equally important. Some philosophers have argued that tracks are what really matter in rock music.

Gracyk claims that ‘rock music is not essentially a performing art, however much time rock musicians spend practicing on their instruments or playing live. And while I do not say that it is essentially a recording art, I do contend that recording is the most characteristic medium of rock’ (1996: 75). Gracyk points out that many rock musicians have been especially interested in creating complex sounds in the studio— that is, in constructing tracks. For example, he quotes John Mellencamp who likes the creativity of crafting songs in the studio but sees stage performance as repetitive (1996: 81). Gracyk notes that this interest in studio craft has led to some rock musicians giving up performance entirely. He writes, ‘Some rock musicians, like the Beatles and Steely Dan, eventually abandon the stage without harming their careers’ (1996: 81). Most musicians do not have the luxury that the Beatles and Steely Dan had. If Gracyk is right, though, then touring and performance are just economic rather than artistic necessities.

Andrew Kania writes similarly that ‘rock musicians primarily construct tracks’ and thus that tracks ‘are at the center of rock as an art form’ (2006: 404). Yet Kania presses the point further. Because a track is not a performance, a rock song is not a work for performance the way that (for example) a classical symphony is. Rather, the rock song is just what is manifest in the track. The track itself is the work. Thus, Kania argues, ‘rock songs are not works, nor are they for anything in particular’ (2006: 404). Although a rock song can be performed live, it is just a loose specification of a sound that is only fully realized on the track. Kania writes, ‘when listening to a rock track, one does not focus on the thin song manifested in it, nor wonder what another rock band would have done with it; rather, one listens to the track as an entity complete in itself’ (2006: 409). This would give us a different reading of the quotation from Rick Beato discussed above— instead of mixing song and track evaluation, Beato is taking all the features to just be features of the track. To put Kania’s claim in Gracyk’s terms, he insists that rock is essentially a recording art. The record is the work.

The claim that tracks are the important artistic product in rock music minimizes the status of both performances and songs. Michael Rings, who endorses the view, makes that consequence clear when he writes that ‘the central work in rock music is the sound structure captured by the recording… as opposed to a song, a score, or a performance (all of which may function as works in other musical traditions)’ (2013: 56).

Let’s take a closer look at performances and songs.


Even though many rock and pop musicians focus primarily or even exclusively on making tracks, that does not mean performances are never events worthy of aesthetic and artistic appreciation. Most musicians start out performing before they start recording tracks. The fact that tracks can be constructed on a laptop computer in a musician’s bedroom means that some musicians may now jump straight to making tracks, but that is only a recent development. Treating performance as not a work of rock would mean that bands who had not recorded yet had never made rock music— that they had at most done some kind of warm up activity for the real thing.

Moreover, there are subgenres in which performance is the primary focus and where recordings are more documentary than they are constructed tracks.

First, David Goldblatt gives the example of urban vocal groups of the 1950s and 60s. In doo-wop music, he argues, ‘the creation of singers, as well as songs, their rehearsing and performing, was a significant part of the historical process of rock without which many singers and songs never would have made it to a point where they could be packaged and distributed’— that is, singing on street corners necessarily preceded recording tracks (2013: 102). Although a great many black vocal groups started recording in the 1950s— Goldblatt echoes Bill Millar’s estimate of 15,000 such groups— most were ensembles who came to the studio already having performed together (2013: 101).

Second, Christopher Bartel (2017) gives the example of hardcore punk in the early 1980s. Hardcore bands focussed primarily on live performance, and records were more documentary than an occasion to craft a distinct, original sound. Mike Watt of the band the Minutemen says, ‘It was the exact opposite of the big leagues — we didn’t tour to promote records, we put out records to promote tours. Records were like flyers…’ (Blush 2010).

Third, there are jam bands. Bartel writes, ‘Like jazz musicians, jam bands place greater value on improvisation rather than album fidelity while yet remaining broadly within the tradition of rock’ (2017: 152). The Grateful Dead both performed extensively and made no effort to make their live sound echo their studio recordings. Fans of the band show great interest in unofficial live recordings, and the band supported this interest. As of this writing, the Internet Archive hosts 15,986 recordings of Grateful Dead performances. To take one other example, the Dave Matthews band produces studio tracks and their performances on television basically follow the recorded versions. In concert, however, they perform as a jam band— improvising and performing songs differently at each concert. This practice shows that they and their audiences care about performance.


The genre of doo-wop also provides an argument for the importance of songs. More than just a matter of performance, doo-wop exhibits a distinctive kind of songwriting. In the Five Satins’ 1956 hit ‘In the Still of the Night’, Goldblatt points out that the ‘shoo doot ’n shoo be doo’ is part of band member Fred Parris’ composition. The nonsense syllables serve as ‘an instant phonetic souvenir easily sung by most anyone.’ Unlike scat singing in jazz, which involves improvising nonsense sounds, in doo-wop the sounds are ‘built into the fabric of the songs’ (2013: 103). His discussion of the use of nonsense syllables and repetition in doo-wop requires considering the songs themselves.

Songs are also important objects of appreciation in rock and pop music more broadly. As Franklin Bruno notes, ‘a number of rock musicians are frequently credited with excellence or merit as songwriters by listeners and practitioners’ (2013: 67). Although Bob Dylan is a singer/songwriter who figures as a frequent example in Gracyk’s discussion of tracks, Dylan’s songs are worthy of attention in their own right. When Dylan was awarded a 2008 Pulitzer Prize, it was for his ‘lyrical compositions of extraordinary poetic power’— similarly for his 2016 Nobel Prize in Literature.

Gracyk, who focusses on musical features rather than on lyrics, would take the invocation of Dylan’s prizes to be unconvincing. He writes, ‘To be blunt, in rock music lyrics don’t matter very much. Or, to be more precise, they are of limited interest on the printed page, divorced from music’ (1996: 65). Lyrics, he notes, are often written after the instrumental parts are mostly complete. For myself, I care more about lyrics than Gracyk does. In general, different listeners may respond to different things— and the same listener may respond to different things in different cases. When listening to Dylan’s 1964 ‘My Back Pages’, I find the song to be more rewarding than the full sound of the track. For Dylan’s 1965 ‘Like A Rolling Stone’, the opposite. (I won’t stake my life on these assessments, and there is more about Dylan in the next chapter.)

Of course, we usually consider rock songs by comparing renditions rather than by looking at sheet music or written lyrics. Nevertheless, we can see the various versions as expressing the possibilities of the song. As Bruno notes, ‘songs can be judged excellent by listeners who do not find merit or take pleasure in their best-known renditions’ (2013: 67). Someone may enjoy a cover despite disliking the original, on account of how the same song is handled in the two tracks. Conversely, covers which fall short of the original can reveal the complexity of a song— where it seemed simple in the original, that was only because of the great skill involved in its recording and production.

Part of the fascination of covers is that they serve as a laboratory to reveal the limits and potential of a song in this way. Even for a great song which is originally recorded in a great track, the cover can provide a different perspective. Paul Dempsey of the band Something for Kate comments that their goal with a cover is to let the audience ‘hear something differently and see how great it is when a really well-written song can be interpreted in a different way and still stand up’ (JJJ 2021b).

The slipperiness of ‘work’

Where does this leave the question of which of these things is the musical work? The everyday concept and phrase, ‘the work’, is woefully slippery. Discussions of popular music often do not use the word (Horn 2000). Let’s consider three things it might mean.

First, looking at what it means to be a musical work in classical music, we can take a work to be the full specification of what a musical event should sound like. The work, a composition as written in a score, specifies exactly what notes the performer is supposed to play. In rock and pop music, the full specification of the sound is given by the track. This yields the conclusion that the track is the work.

As Lydia Goehr argues, understanding works and performances in this sense just reifies ‘ideals that exist within classical music practice’ (2007: 99). Since about 1800, classical music has been guided by an ideal of ‘perfect compliance’— that the performer should play all and only the notes in the score. Goehr points out that this norm is a distinctive development. She writes, ‘it is significantly this ideal that serves to distinguish the practice of producing performances of classical music works from the performance practices associated with other kinds of music’ (2007: 99). Although it is typical to hear rock and pop songs as tracks, the discussion of performances and songs above shows that tracks do not (or at least do not always) play the same regulative role that compositions do in classical practice. So this sense of ‘work’ will not underwrite an argument about rock music in general.

Second, we can take the work to be the durable object. This marks the distinction between works (as repeatable) and performances (as fleeting). Without recording, each performance is a one-time event while the work could be performed many times. The score for the work is a tangible thing. It is a product or commodity which can be stocked on a store shelf, held as inventory, and sold. With recording, the record is the tangible thing. Even though digital files have replaced physical media, the files too can be stored and traded.

This is a difference in the metaphysics, what kind of thing each of them is. Yet it has consequences for our appreciative practices. The fact that an unrecorded performance is a singular event whereas a track may be listened to many times makes a difference in listening (Davies 2001). However, a recording of a performance might give a listener access to all the appreciatively relevant features and can also be listened to repeatedly (Mag Uidhir 2007, Magnus 2008). So this metaphysical distinction does not establish the exclusive primacy of tracks.

Third, we can take the work to be the object of artistic appreciation— as Charles Nussbaum puts it, ‘the object of proper musical regard’ (2021: 329). Gracyk points to ‘the works that the artist sanctions as items for appreciation and critical evaluation’, thus distinguishing works from sketches and preparatory drafts (1996: 35). Because (many) musicians intend for audiences to primarily appreciate their published tracks, this yields the conclusion that the track is the primary work in rock music.

As examples of doo-wop and punk show, however, not all rock musicians take tracks to be their primary focus. Even if they did, it is not clear that such an intention would proscribe appreciating a song or performance.

Imagine a luthier who is very proud of the guitars they make. They record themselves playing one of their guitars. Because it is intended to help them sell guitars, the recording is meant to be appreciated and evaluated as documentation of how their guitars sound. Even so, I do not see anything untoward about a listener appreciating the performance itself. You cannot assume that the guitar will sound that good if you buy one and play it yourself, but that just underscores that appreciating the performance— although permissible— is different than the intended appreciation of the instrument.

In much the same way, people do appreciate songs and performances even from artists who see tracks as their primary product. And there is nothing wrong with that.

Ultimately, the question of which of the three is the work is misconceived. They are all important parts of musical practice, both for artists and audiences. As Bartel suggests, ‘we should see rock as a tradition that has three activities at its core: songwriting, live performance, and track construction’ (2017: 153). He calls this view ‘rock as a three-value tradition’, and Dan Burkett calls it ‘a pluralist ontology of rock’ (2015).

One might object to this pluralism on grounds of parsimony or simplicity. The idea is that Ockham’s razor should stop us from positing many works rather than one. Ockham’s razor is the principle that entities should not be posited any more than necessary, but the objection would misuse the principle in at least two respects. First, recognizing all three categories does not require recognizing any extra entities. Songs, performances, and tracks all exist, regardless of which are works. Second, the fact that important parts of musical and critical practice look to songs and performances means that recognizing them does not multiply distinctions beyond necessity. Rather, it is exactly what is necessary to make sense of these practices.


Gracyk and others are right to insist that tracks are importantly different works than songs. Especially before Gracyk, authors often failed to distinguish them. So pluralism acknowledges Gracyk’s greatest contribution.

As I noted earlier, distinguishing these three kinds of things as songs, performances, and tracks is an explication. Since they mark useful distinctions, I will follow that usage as much as possible in this book. I will use the word ‘song’ to indicate the repeatable thing which loosely specifies what the lyrics and tune should be, the word ‘performance’ to indicate either the live event of playing a song or a recording that documents such an event, and the word ‘track’ to mean a recording constructed in the studio.

A few other terms will also be useful. When the distinction between a recorded performance and a track is not relevant, I will use the word ‘recording.’ When the distinction between performances and tracks is not especially relevant, I will use the word ‘version’ to mean a performance or recording of a song.

What is it that a cover is a cover of?

The explications above make more precise the condition (from the end of the last chapter) that a cover must be a musical version which covers an earlier recording. Covering is a historical relation between an earlier recording and a later version of a song. The earlier version is taken as the canonical one, and the cover is based on it. It is the target of the cover, the source for the new version. Kurt Mosser calls it the ‘base’ version, the one ‘that, due to its status, popularity, or possibly other reasons, is taken to be paradigmatic’ (2008).

The target of the cover is often the original or first-published recording of the song, but not always. For example, the Beatles’ version of ‘Twist and Shout’ was chosen by Rolling Stone readers as one of the greatest cover songs of all time in a 2011 poll. In that article and in many other places, it is described as a cover of the Isley Brothers’ 1962 track even though the first release of the song was by the Top Notes in 1961.

Kania argues that covers are not indexed to earlier recordings in this way. He argues instead that ‘covers can be grouped together as tracks intended to manifest the same song’ (2006: 410). For example, we can organize the Beatles’ track, the Isley Brothers’, the Top Notes’, and dozens of others besides as versions of ‘Twist and Shout.’ Although it is true that we can group them together that way, it leaves out important information about inspiration and influence. Calling the Beatles’ version a cover of the Isley Brothers suggests that they had it in mind. The sound of the Beatles’ track reflects their choices and abilities in relation to the version that inspired them. We can credit the cover for improving on the original or fault it for falling short. Contrawise, it would make no sense to think of the original as influenced by the cover. Just considering all the versions of ‘Twist and Shout’ together would overlook this asymmetry.

Still, we should consider the two reasons Kania offers to support his claim that a cover is indexed to a song rather than to an earlier version.

First, Kania maintains that appreciating a cover does not require considering the earlier track of which it is a cover. He writes, ‘although a band may take just one version of a song as their target, knowledge of this does not seem relevant to critical assessment of their track’ (2006: 411). He argues for this by example, providing an extended discussion of the Pet Shop Boys’ 1987 hit cover of ‘Always on My Mind’— a song made famous by Elvis Presley in 1972. Kania writes, ‘Willie Nelson covered “Always on My Mind” in 1982, between the Elvis and Pet Shop Boys versions. Both of the later versions are covers of the same song. It would not make any difference to this situation if the Pet Shop Boys had never heard Elvis’s track and only intended to cover Nelson’s’ (2005: 411). Discussing the Pet Shop Boys’ cover, Kania notes that the chord progressions and arrangement are more complex than in Elvis’ version. By their own declaration, ‘the Pet Shop Boys wanted to construct a track that sounded as different from Elvis’s as possible’ (2006: 410). So it is unsurprising that critics and audiences evaluate the cover in relation to Elvis’ track, rather than (as Kania would suggest) just evaluating it as a version of the song. For example, The Telegraph eulogizes that the cover ‘elevated Elvis’s tender elegy… into a monumental explosion of high pop camp’ (2004).

Also consider Bob Dylan’s 1967 ‘All Along the Watchtower.’ It was covered the following year by Jimi Hendrix, and the musical changes introduced in Hendrix’s version became iconic. As Janet Gezari and Charles Hartman write, Dylan ‘adopted Hendrix’s stylistic take on his song, as revealed in many live recordings in the seventies and after. In effect, he covered a cover of his own song…’ (2010: 165). To understand what Gezari and Hartman as saying, one needs to think of Dylan’s later versions not merely as instances of the song (which he wrote) but as being in relation to a version which they are covering (that is, Hendrix’s recording).

In both of these cases, critical assessment of the cover turns on considering it in relation to an earlier track taken as a canonical original. However, in neither case is the canonical original the first recording of the song— Gwen McRae and Brenda Lee both released recordings of ‘Always on My Mind’ before Elvis, and Dylan himself had first released ‘All Along the Watchtower.’

Second, Kania writes that ‘rock audiences seem to group covers with respect to the song they are intended to manifest, rather than simply by the track(s) taken as the immediate object of the covering intention. Covers do not only occur paired one-to-one with originals’ (2006: 411). He is right that the covering relation— although typically a one-to-one relation between a cover and an original— is not always so simple.

The case of ‘Respect’ illustrates how the relation can be history-relative and audience-relative. Aretha Franklin’s 1967 hit version was a cover of an Otis Redding hit from two years prior. Franklin changes around the lyrics and narrative of Redding’s song, and some of the most memorable lyrics are Franklin’s addition. She adds spelling out the word ‘R-E-S-P-E-C-T’ and saying ‘Find out what it means to me.’ Although Franklin’s version was not the first, it can be taken as canonical and as the target of a cover. Gracyk writes, ‘It’s easy to forget that Aretha Franklin’s titanic, defiant version of “Respect” (1967) is a cover version of one of Otis Redding’s 1965 hits, and her action of appropriating “his” song was originally an important element of its affective power’ (2001: 209). Kelly Clarkson performed ‘Respect’ in the first season of American Idol in 2002, and Clarkson’s version is labelled as an Aretha Franklin cover. We can only understand this shift by thinking of a cover as being of a track which is taken as canonical, rather than as being merely another instance of the same song. Clarkson takes Franklin’s version as canonical both by having it in mind and by singing Franklin’s additional lyrics. We would evaluate Clarkson’s version differently if— impossibly— she had never heard Franklin’s version but had just made up for herself the lyrics which differed from Redding’s.

Moreover, the relation can be one-to-many instead of one-to-one. Consider college a cappella choir versions of Dr. Dre’s ‘Bitches Ain’t Shit’; for example, by DeCadence at UC Berkeley in 2010 and Continuum at Eastern Michigan University in 2011. The song originally appeared on Dre’s 1992 album The Chronic, but Ben Folds had a hit cover of it in 2005. Folds’ cover is a piano and voice arrangement that transforms Dre’s rap into a serenade, and he has sometimes performed it a cappella. The various vocal groups who have performed it follow Folds’ version, adapting it for choir. Audiences were likely to be familiar with both the Dre and Folds versions, so we might sensibly say that the choir performances are covers of both hit tracks.

So the relation can be more complicated than just pointing to one earlier track which counts as the canonical original for all time. Nevertheless, there is always a musical-historical relation that connects a cover to one or more earlier recordings. Absent that relation, it is not a cover at all but just a version of the song in question.

Mimics and renditions

Of course, there are different possible relations that a cover can have to its target original.

Some covers aim to echo the interpretation and arrangement of the canonical recording precisely. Call these mimic covers. A perfect mimic cover would be indistinguishable from the original.

This means that a recorded mimic cover is, in a sense, redundant. One could always just listen to the original instead. Even though one can easily go listen to the original track, it is not so easy to attend a performance by the original artist. A cover band might be the only option. The result is that most mimic covers are performed live. Mosser presses this further, writing that ‘these kinds of covers are invariably live performances’ (2008: §II.a).

However, there are recordings of mimic covers. Consider two kinds of examples.

First, mimic covers can be technical exercises. Zak describes a mimic cover recording by Karl Wallinger: ‘Much as a composer of scores might copy out a favorite piece by another composer, Wallinger remade “Penny Lane” note for note and sound for sound, including reverb.’ The point is not that anyone else would want to listen to Wallinger’s mimic track. Rather, Zak explains, ‘What he learns from such an undertaking he uses in making his own records, which in turn bear the sonic influence of, among other things, the later Beatles albums he is partial to’ (2001: 36). This technical exercise is like a jazz student transcribing and performing a mimic of a famous jazz solo. The point of such an exercise is not to entertain an audience, but instead to develop skills and a vocabulary of sounds that can be used later in original undertakings.

Second, mimic covers can be jackal tracks meant to cash in on the success of the original. The Typhoons were the house band at Embassy Records that recorded the label’s mimic covers of Beatles’ tracks. They covered dozens of Beatles songs from 1962 to 1965, but did nothing outside the studio. Somewhat later, from 1968 to 1985, there was a record series called Top of the Pops produced by Hallmark. Each of the albums in the series was a collection of recent songs recorded by in-house artists and sold at a low price. In the mid-1970s, one of these even topped the best-selling album charts. People did not buy it for those versions especially, but because it was cheap and readily available. (The rules for what was counted in the charts were changed so that this did not happen again.) Later still, there are artists who post mimic covers to electronic music services in hopes of profiting from accidental clicks. All these mimic covers are meant for an audience, although an audience that only listens to them for reasons of economy or misdirection.

The Top of the Pops record series comprised albums of mimic covers put together on the cheap. Even the album art was recycled. (a) A 1966 publisher sampler album from the U.S. (b) The first Top of the Pops album, released in 1968, reuses the same photograph. It was used again as part of the album cover for another Top of the Pops release the next year. (Images used with permission of the Pickwick Group.)

As Gaynor Jones and Jay Rahn suggest, the creation of mimic covers is ‘closer to classical performance practices than to those which we usually associate with popular music’ (1977: 85). A pianist playing a Beethoven piano concerto typically tries to play what is written in the score, note for note. In a similar way, someone performing a mimic cover takes the original track as specifying what a version of the song ought to sound like. They attempt to follow the original track, sound for sound. In a certain respect, this is even more restrictive than the practice in classical music. Practices in classical music allow ‘narrow decisions of dynamics, phrasing and articulation’ (Hamm 1994: 149). As Zak notes, there has been a trend of narrowing these decisions even further; he writes that ‘classical compositions of the nineteenth and twentieth centuries have generally tended toward increasingly detailed written specifications, which attempt to define a work’s parameters with ever greater precision’ (2001: 42). This is what Goehr (2007) calls the ideal of ‘perfect compliance.’ Tracks, ‘like scores, are detailed and fixed representations of musical thought’ (Zak 2001: 42).

Other covers do not treat the original in this way and do not aim simply to echo features of the original. Call these rendition covers. Most recorded covers are renditions, where the recording artist or band is not attempting to fully emulate or impersonate the original. Some renditions are fairly straight, mostly following the lyrics and musical choices of the original but involving some changes in instrumentation and sound that reflect the abilities and interests of the musicians recording the cover. Others are more transformative, like Hendrix’s cover of ‘All Along the Watchtower’ and Franklin’s cover of ‘Respect.’

It is common to suggest that covers exist on a continuum. For example, Mosser distinguishes different types of covers and writes that ‘if we take “cover song” as a genus, we can identify a continuum of species that ranges from an attempt to reduplicate a given song or performance to a parody that maintains only the most minimal connection with the song being parodied’ (2008). Deena Weinstein writes similarly, ‘Covers exist along a continuum from those that are drastically different from, to those nearly identical to the original’ (2010: 245).

This treats the types of covers as a matter of degree. That is, the difference is just how much the cover sounds like the original. However, a mimic cover need not and generally will not sound exactly like the original. A poorly executed mimic sounds quite different, and a fairly straight rendition can sound more like the original than a botched mimic does. What makes one a mimic and the other a rendition is their context of creation and appreciation. As I argue in the next chapter, the standards for what counts as a good cover are different for mimic covers than for rendition covers.

Moreover, thinking in terms of a continuum invites thinking of differences just along one dimension. This is plausible for a mimic cover, where any difference from the original is a demerit. For a rendition cover, however, differences can occur along many different dimensions. Changes to all the particular notes may preserve the harmonic progression. Conversely, a change in a few notes can make for a change from major to minor and thus change the whole mood. Changes to almost all the words in the lyrics might convey the same overall meaning, or a different change to just a few words could result in a substantial change to the meaning.

So placing covers on a continuum based on resemblance to the original fails to capture the distinction between mimic covers and rendition covers, and it smooths over important factors in the appreciation of rendition covers.


I have argued that understanding recent music better requires distinguishing songs, performances, and tracks. A cover is a performance or recording which covers an earlier recording. I have also argued that we should distinguish mimic covers from rendition covers— this distinction is put to more work in the next chapter.

Powered by Epublius