MAGNUS Et Al-2013-The Journal of Aesthetics and Art Criticism

Embed Size (px)

DESCRIPTION

Descriptive art

Citation preview

  • CRISTYNMAGNUS, P.D. MAGNUS, AND CHRISTY MAG UIDHIR1

    Judging Covers

    abstractCover versions form a loose but identifiable category of tracks and performances. We distinguish four kinds of covers andargue that they mark important differences in the modes of evaluation that are possible or appropriate for each: mimiccovers, which aim merely to echo the canonical track; rendition covers, which change the sound of the canonical track;transformative covers, which diverge so much as to instantiate a distinct, albeit derivative song; and referential covers, whichnot only instantiate a distinct song, but for which the new song is in part about the original song. In order to allow for the verypossibility of transformative and referential covers, we argue that a cover is characterized by relation to a canonical trackrather than merely by being a new instance of a song that had been recorded previously.

    Common usage distinguishes originals from cov-ers. Roughly, the distinction is this: An original isa recording or performance of a song either by theperson who wrote it or by the person for whom itwas written. A cover is a later recording or perfor-mance by someone else.

    We argue for a number of related claims: First,there are different kinds of covers correspondingto different evaluative criteria that are applicableand appropriate to them (Sections IIV). Second,a recording or performance counts as a cover rela-tive to a canonical version. So its status as a coveris importantly a situated, historical fact (SectionIV). Third, a cover may differ from canonical ver-sions in ways that make it an instance of a dis-tinct, albeit derivative song (Section IV). Fourth,in some cases a cover can take on a novel mean-ing such that the new, derivative song is about thecanonical one (Section V).

    None of our arguments appeal to the preciseessence of covers as such. We are not concernedwith fundamental metaphysics. Rather, we takecovers as they are identified in commonusage, andour claims are compatible with different accountsof the underlying ontology.

    i. what is a cover?

    The phrase cover version originated in the 1960sto describe a new recording or performance of a

    song that someone else had already made a hit.A typical dictionary definition identifies a coveras a recording of a song, etc., which has alreadybeen recorded by someone else.2

    This definition is too simple and includescases which one would hesitate to call covers.For example, Patsy Clines 1961 recording ofCrazy is taken as the paradigm version of thesong. Rolling Stone identifies Clines version asthe eighty-fifth greatest song of all time.3 Yet thesong was written by Willie Nelson, and Cline firstencountered it when she heard Nelsons recordeddemo of the song. Merely that it was recorded bysomeone else, as on the demo selling the song,is not enough to make Clines track a cover. Onecould require a published recording, which wouldexclude Nelsons demo and preserve the intuitionthat Clines version is not a cover. But Nelsonlater recorded his own version of Crazy, and it isvexedwhether his version counts as a cover or not.

    If we were concerned with the semantics ormetaphysics of covers as such, we could revisethe definition further. By an iterative process ofcounterexample and monster-barring, our defini-tion would become more cumbersomebut nomore enlightening. Our intuitions are probablynot strong enough to survive such a vigorousprocess.

    Gabriel Solis argues that covers both de-fine and are defined relative to rock music. He

    The Journal of Aesthetics and Art Criticism 71:4 Fall 2013C 2013 The American Society for Aesthetics

  • 362 The Journal of Aesthetics and Art Criticism

    understands rock to be rock-as-tradition, whichgroups together hard rock, punk, hardcore,metal, and so on, and he defines covers in termsof the versioning practices of that tradition.4 Sucha definition substitutes the shadowy for the ob-scure, because rock-as-tradition is itself nebulous.It is worse perhaps, because rock in this broadsense is not a term of ordinary language. As such,we cannot rely on common usage and intuitionsabout cases.

    We do not hope to provide a precise definitionof cover. We are interested primarily in issues ofevaluation rather than definition, so it suffices thatwe begin with rough agreement about some per-formances and recordings that are cover versions.To begin, we take the following to be a prima faciesufficient condition for being a cover: A versionof a song is a cover when it is recorded or per-formed by an artist or a group who did not writeand compose the song themselves andwhere thereis a prior recording which is accepted as canonicalor paradigmatic.

    Because this is only a sufficient condition, thereare some acknowledged cover versions which donot fall under it. We return to this point below(Section IV).

    The condition relies on the notion of a canoni-cal or paradigmatic recording, what Kurt Mossercalls the base version of the song.5 A versionsstatus as canonical is clearly something that canvary over time. It is not entirely listener relative,but it is relative to an audience or a community. Itis possible for an artist to record a cover version ofsome song and for her cover to become the canon-ical version; in later contexts, other versions maybe covers of it. (We give examples further on.)

    In what follows, we consider only works of pop-ular music with lyrical content. We are neutral onthe question of whether covers, strictly speaking,are only possible in the rock tradition, but we willavoid examples from beyond it.

    To reiterate, we are not presuming that cover isa sharp metaphysical category. However, we aremaking some ontological assumptions. We mini-mally assume distinctions between songs, tracks,and performances:

    1. Lyrics and musical structure, thinly specified,provide the identity conditions for a song.Interpretation allows for performances andrecordings to be of the same song even if theysound different.6

    2. A track or recording is a complete specifica-tion of sounds. A track typically involves ele-ments from several sessions and a great deal ofwork in the studio, and it is something besidesa performance. The track can exist as multiplecopies and in different media, but any differ-ences in sound are deficiencies.7

    3. A performance is connected to a dated event.The dated event can only happen once, and alater performance would be a distinct instanceof the song.8

    We do notmake any strong assumptions about thefundamental nature of songs, tracks, and perfor-mances. One might interpret them in terms of aninflationary ontology according towhich songs areuniversals, recordings are subordinate universals,andperformances are particulars.Alternately, onemight prefer a deflationary ontology according towhichwhat strictly speaking exists are sonic eventsthat can stand in the same-song or same-track re-lation to one another. Or one might treat songsand tracks as historical individuals.9 We merelyinsist on the possibility of reidentifying the samesong, and any plausible ontology must allow forsuch a commonplace activity.

    ii. mimic covers

    Some covers merely aim to echo the arrangementand interpretation of the canonical version. Callthese mimic covers. Mosser calls these reduplica-tion covers and suggests that one might deny thatthey are covers at all.10 This reticence strikes us asodd. For one thing, mimic covers fit the sufficientcriterion that we formulated in Section I. For an-other, bands that play mimic covers exclusivelysometimes called tribute bandsare also com-monly called cover bands. Mosser himself givesthe example of the Dark Star Orchestra, a groupthat aims to faithfully reproduce performancesby the Grateful Dead. Other examples includeFan Halen (a Van Halen cover band), BC/DC(an AC/DC cover band from British Columbia),and the Iron Maidens (an all-female Iron Maidencover band).

    Few fans would opt to see the tribute band, allthings being equal, but the original band may nolonger exist, may not perform locally, and will notbe performing all the time. So cover bands pro-vide a surrogate for seeing a live performance by

  • Magnus, Magnus, and Mag Uidhir Judging Covers 363

    the canonical band. A maximally successful coverbandwould be acoustically indistinguishable fromthe canonical band.A striking example is the bandBritish Steel, a Judas Priest cover band. WhenRob Halford, the original lead singer of JudasPriest, left the band, Tim Ripper Owens, thelead singer of British Steel,was recruited to re-place him. Owens was selected precisely becausehe could sing like Halford.11

    What holds for the oeuvre of a cover band alsoholds for a single mimic cover: the best mimiccover sounds just like the original.

    Suppose we try to evaluate the mimic cover bysome standard other than fidelity to the canon-ical versionconsider, for example, the song orthe interpretation instantiated in the mimic cover.Because the aim of the mimic cover is to echothe canonical version, the two are instances of thesame song and the same interpretation. Insofar asthe mimic cover differs, it is a defect of craft. So,for purposes of evaluating the song or interpre-tation, listening to the mimic cover is like listen-ing to the canonical version being played back.Failures of the mimic cover are like deficiencies inthe playback, and these may hinder our ability tofairly evaluate the song or interpretation. All thesame, we are evaluating the canonical version andnot the mimic cover.

    Mimic covers are typically performed live.12

    Since the primary mode of evaluation for mimiccovers is fidelity to the original, it is no surprisethat studio recordings of mimic covers are rare.A perfectly successful mimic cover track wouldbe indistinguishable from the canonical track. Yetmimic cover tracks are possible. A friend of oursfrom grad schoolan amateur, aspiring recordingengineerrecorded his own versions of severalpop tracks. Other friends recorded instrumentsand voices, and he worked to have the productionand effects match the canonical version. For him,success was matching the sound of the original.These recordings were technical exercises. Eval-uated as works of art, they were redundant. Heintended themmerely to beworks of craft, demon-strations of and practice for his audio productionskills.

    iii. renditions

    Tribute bands perform covers exclusively, butmany other performers include one or two covers

    as part of a set that is otherwise their own orig-inal songs. Consider the singer-songwriter VanceGilbert performing the Jimi Hendrix song Cas-tles Made of Sand. Gilbert introduces his ownsongs with stories about his life, but he introducesCastles Made of Sand simply by saying that itis time for a little Hendrix. Gilbert plays it onan acoustic guitar, rather than an electric, andsings it with the same voice he uses for his owncompositions.

    Gilberts cover is not a mimic cover. He is notattempting to sound like Hendrix. Although itis clearly an instance of the same song as thecanonical version, the performance is not meantto sound like the Hendrix track. While acknowl-edging the canonical version, Gilbert is playinghis own rendition of the song. Call this a renditioncover.

    Rendition covers, unlike mimic covers, are of-ten recorded tracks. Such recordings are not re-dundant in the same way that recorded mimiccovers are, because the covering artist or band hasmade different interpretive choices. For example,the Swedish rock band TheCardigans has covereda number of songs by the heavy metal band BlackSabbath. TheCardigans cover of IronMan tellsthe same story using the same words as BlackSabbaths canonical version. Yet one could hardlymistake one version for the other. The emotionalcontent is differentone might find the faster,louder canonical version to be more threaten-ing and the Cardiganss softer cover to be moresad. If we treated the song as thickly specified,this difference would make the Cardigans IronMan a different, although derivative, song thanthe original. Following Andrew Kania, we takerock songs to be thinly specified: different inter-pretive choices are compatible with tracks or per-formances being of the same song. This allowsus to acknowledge the straightforward sense inwhich the Cardigans recorded a Black Sabbathsong.13

    Songs being thinly specified allows not justfor some musical differences but also for somechanges to the lyrics. For example, on the 1993 al-bum Across the Borderline, Willie Nelson sings acover of Paul Simons Graceland. There is a linein the song in which Simon sings, There is a girl inNew York City, who calls herself the human tram-poline; Nelson sings it as a girl inAustin, Texas.The change from NYC to Austin fits Simons nar-rative while fitting better with Nelsons idiom. So

  • 364 The Journal of Aesthetics and Art Criticism

    Nelson is playing the same song as Simon eventhough he has changed some of the lyrics. Notethat this flexibility is not limitless, even if the lim-its are vague. If Nelson had substituted a differentdestination for the title attraction, then he wouldnot have been singing Graceland.14

    These examples illustrate that rendition coversoften involve a change in musical genre. Yet theyneed not. The Bangles 1980s cover of Simon andGarfunkels 1960s Hazy Shade of Winter, forexample, is a rock song played as a rock song. Yetit is not just a mimic cover. Tracy Hoff of the Ban-gles recounts, When I heard that song, I thought,thats so perfect for The Bangles.15 They beganplaying it in concert, and their arrangement andinterpretation made it sound like a Bangles song.Bothmetaphorically and literally, they played it intheir own voice. The Bangles studio track of thesong, recorded for a movie soundtrack, becamea hit.

    We should not evaluate rendition covers bytheir sonic fidelity to the canonical version. In-sofar as we have the canonical version in mind,we attend primarily to the ways that the rendi-tion cover departs from it. We compare a mimiccover to the canonical version, but we contrast arendition cover with it. A rendition cover, whenconsidered in relation to the original, is successfulin this way insofar as it departs from the canonicalversion in artistically interesting or virtuous ways,a failure to the extent that it departs in artisticallyuninteresting or defective ways, and prima facieworthwhile only to the extent that it departs at all.

    This mode of evaluation, which focuses on re-spects in which the cover diverges from the canon-ical version, is especially important for assessingcreativity and for assigning credit. The Cardiganscover of IronMan is a tour de force because it isshockingly different from the Black Sabbath ver-sion, and the band deserves credit for imaginingit. Contrast recent a cappella versions of Dr. DresBitches Aint Shit, including ones performedby college a cappella groups at Eastern MichiganUniversity andUCBerkeley.16 The song originallyappeared on Dres 1992 album The Chronic, butBen Folds had a hit with a cover of it in 2005.Foldss cover is a piano and voice arrangementthat transforms Dres rap into a serenade, and hehas sometimes performed it a cappella. The var-ious vocal groups who have performed it followFoldss arrangement. The groups deserve creditfor a successful performance and perhaps for ar-

    ranging several voices, but credit for the surprisinggenre shift goes to Folds.17

    Musicians may choose to perform or recorda cover as a way of acknowledging the canoni-cal version. As numerous authors note, a covermay show an important influence on the artist.18

    Mosser gives this example: Talking Heads coverof Al Greens Take Me to the River . . . tells. . . the Talking Heads audience that the bandhas listened not just to Al Green, but more likelythan not to a good bit of Rhythm and Blues;while not part of that tradition, its references toit help establish its musical bona fides and em-phasize the importance of those sources.19 How-ever, one should not take this point too far. A fanwho knows nothing of Al Green learns nothingabout the Talking Heads musical tastes. Rather,the song is just one they perform but did not write.Even a fan who knows about Al Green will not beable to infer much from the cover. Perhaps theTalking Heads heard this one R&B song and de-cided to cover it, as the Bangles did with HazyShade of Winter. Perhaps they only covered thesong because their producer recommended it. In-sofar as the cover points to influences on theband, it does so in combination with other facts.20

    These associations are relevant to an assessmentof the band, perhaps. We may even consider themwhen assessing the cover as a historically situatedwork.

    Further context beyond just knowing about thecanonical version is required to assign credit andassess the creativity of a rendition cover. It mayalso be important to consider the cover in con-trast with other covers of the same song, to othersongs in the oeuvre, or to other songs by con-temporary artists. Indeed, a new song can failto be original if it is utterly typical of its genre.A performance of such a dud, although not acover, would be stultifying and banal. A covermay fail to be original for similar reasons, evenwhile differing markedly from the canonical trackwhich it covers. Assessing novelty and creativ-ity requires considering context, and what a per-formance or track covers is part of a coverscontext.

    According toMichael Rings, the chief appeal ofgenre-shifted renditions is that they can surpriseus by treating familiar material in an unfamiliarway. My proposal, Rings writes, is that muchof the pleasure in listening to these covers comesfrom following a familiar songs progress through

  • Magnus, Magnus, and Mag Uidhir Judging Covers 365

    an unfamiliar (relative to the song) stylistic land-scape: how will this next passage sound within theconventions of this new genre?21 Of course, whenone evaluates a genre-shifted cover in relation tothe canonical version, one can attend to differ-ences in how the song sounds in its new setting.However, it does not follow that this is the only, oreven necessarily the predominant, interest of suchcovers. First, one can only sincerely wonder howthe renditionwill go the first time that one hears it.On subsequent listenings, one remembers how itwent. If anticipation and surprisewere theprimaryappeal of genre-shifted rendition covers, then theyshould stop being interesting after one had heardthem enough to be familiar with thembut many,like the Cardigans Black Sabbath covers, do not.Second, there is a way in which rendition coverscan be aesthetically interesting when there is nochange in genre. Surely at least some genre-shiftedrenditions are rewarding in that way, rather thanin virtue of the genre shift itself.

    In contrast with evaluating a rendition coverin relation to the canonical version, there is analternate mode of evaluation we can take: consid-ering it on its own merits. We can assess the songand the interpretation of the song that are instan-tiated in the cover without considering whetherit is the performer or someone else who wrotethe song and made the interpretive choices. Ka-nia points to the possibility of such evaluationwhen he writes that although a band may takejust one version of a song as their target, knowl-edge of this does not seem relevant to criticalassessment of their track.22 This can clearly di-verge from the first mode of evaluation. To of-fer a schematic case, imagine a tedious song writ-ten by singer-songwriter A who makes an artlessrecording, awkwardly fumbling through the songwhile playing guitar. Later, group B competentlyarranges it for a full band and skillfully performsa rendition cover. In the first mode of evaluation,we can judge the cover to be an artistic success.B has departed from As original in a rewardingway. In the second mode of evaluation, we canjudge the cover to be a failure. The resulting per-formance might still be tedious because the songitself is just that bad.23

    Regardless of how we describe the difference,there are clearly these twodifferentmodes of eval-uation that canbe employedwith rendition covers:the former considers it relative to the canonicalversion; the latter does not.

    iv. transformative covers

    One might think that mimic covers and renditioncovers would exhaust the possibilities: covers in-tended to sound like the canonical track and thoseintended to sound different, respectively. We ar-gue that this is not so, that there are further im-portant distinctions to be drawn.

    Consider the case of Aretha Franklins classicRespect, a hit in 1967. Many listeners think ofit as Franklins song. Yet it is a cover of a 1965Otis Redding hit of the same name. Common us-age recognizes Franklins track as a cover. For ex-ample, The Telegraph ranks it as the fifth great-est cover ever recorded, and Rolling Stone ranksit as the fifth greatest song of all time, calling itthe definitive cover.24 Nevertheless, FranklinsRespect is not merely a rendition of Otis Red-dings song. Her version adds lyrics absent fromthe original: spelling out the word R-E-S-P-E-C-T, the change from give me my proper re-spect to give me my propers, adding the lineTake careTCB (meaning, take careof business),adding the line and backup vocal Sock it to me,and so on. Beyond these additions, Franklin alsochanged some of the original lyrics. Where Red-ding sang, All Im asking is for a little respectwhen I come home, Franklin sang, All Im ask-ing is for a little respect when you get home.This transforms Reddings ultimatum to a house-bound woman into a demand for consideration,one which might be made between equals. TheRolling Stone article describes this as the differ-ence between a brawny march and calling anend to the exhaustion and sacrifice of a raw deal.These differences go well beyond the change toa few chord progressions or the substitution of afew inconsequential words. This is not merely adifference in musical style or idiom, but a signif-icant difference in content. So Franklins track isbest understood as instantiating a distinct, thoughderivative, song from Reddings, even treating hiscomposition as a thinly specified rock song.

    Understood in this way, Franklins Respectwill not count as a cover according to the crite-rion we offered in Section I. That was only of-fered as a sufficient condition, however. We canadd this alternate sufficient condition: A track (orperformance) is a cover when it is of a song that issufficiently derivative of a song that the artist orgroup did not write themselves, and where thereis a recording of the original song that is accepted

  • 366 The Journal of Aesthetics and Art Criticism

    as canonical or paradigmatic. Call a cover thatsatisfies this alternate condition a transformativecover.

    We mean derivative in a noncritical, geneticsense, rather than as a synonym for banal. Howderivative a songmust be inorder tobe sufficientlyderivative is admittedly vague, but common usagesuffices to draw a rough boundary. Franklins Re-spect is sufficiently derivative of Reddings, andcommon usage identifies it as a cover. Weird AlYankovics Yoda is not sufficiently derivativeof the Kinks Lolathe two share no lyrics atalland common usage identifies it as a parodyrather than a cover. There may be some unclearboundary cases, but the rough distinction will beclear enough for the cases we consider in this arti-cle. (Recall that our primary interest is evaluative.Vagueness would be a greater worry if our projectwere metaphysical.)

    Andrew Kania gives an account according towhich a cover must be an instance of the verysame song as the canonical track. Moreover, onhis account, the object of a cover (what is coveredby it) is the singular song rather than the originaltrack. So Kania would reject the very possibilityof a transformative cover.

    Kanias case depends partly on his choices ofexamples. He provides a lengthy discussion of thePet ShopBoys cover ofElvis Presleys Always onMy Mind. The cover was deliberately arrangedso as to be different from the Elvis track. For ex-ample, the chord progressions are more complex.Kania uses the case to illustrate the importance ofthe track as an object of aesthetic attention; the de-liberate effort not to sound like Elvis shows whatthe Pet Shop Boys want their track to be mea-sured against . . . Elviss recording of the song,rather than the thin song itself.25 The connec-tion between Elviss version and the cover is es-pecially salient because the cover was recorded aspart of a tribute to Elvis, and the connection isstrong enough that Kania does not even mentionthe pre-Elvis hit recording of the song by BrendaLee. Kaniamaintains (correctly, we think) that thePet Shop Boys cover is an instance of the samesong that is instantiated by Elviss track. Yet heconcludes from this (wrongly) that a cover mustalways instantiate the very same song as the priorversion and so, further, that the object of the coveris just the song and not the prior version at all.

    Since the Pet Shop Boys track is the same songas the Elvis track but is intended to sound dif-

    ferent, it is a rendition cover. As such, there aretwo possible modes of evaluation. Engaging inwhat we identified as the first mode of evaluation,The Telegraph eulogizes that the Pet Shop Boysversion elevated Elviss tender elegy . . . intoa monumental explosion of high pop camp andlauds it as one of the best covers ever recorded.26

    This mode of evaluation requires identifying thetrack not merely as a cover of Always on MyMind but as a cover of the Elvis track. So wethink it makesmore sense to say (as we did above)that the object of the cover is a prior track ratherthan merely the song. Kania argues against such amove on the grounds that rock audiences seemto group covers with respect to the song they areintended to manifest, rather than simply by thetrack(s) taken as the immediate object of the cov-ering intention.27 Yet he offers no reason to thinkthis is so, and actual usage suggests the opposite.

    Returning to the example of Franklins Re-spect, Kanias account would force us to say ei-ther that the Franklin track is not a cover (in oppo-sition to commonusage) or that she sings preciselythe same song that Redding composed (in oppo-sition to common sense). The case of Respectposes what he calls the striking cover paradox, aproblem he imagines only in the abstract.28 Kaniamight accept the second horn of the dilemma hereand say that Franklin really did just sing the samesong that Redding wrote, admitting just that shesang it in an original way. Such insistence wouldstrike us as implausible. All of the memorable,quotable bits of Franklins Respect are absentfrom the Redding version: the spelled-out R-E-S-P-E-C-T, my propers, and sock it to meand also Find out what it means tome.Aswe ar-gued above, the different lyrics make a significantdifference to the meaning. Redding and Franklinboth sing about respect, but they say importantlydifferent things about it.

    The caseofFranklins Respect illustrates howcover is a history-relative and audience-relativenotion. Her version quickly became the canonicalversion of the song. Although a derivative work, itis much richer than the original. It resonates withstruggles for equality and civil rights in ways thatReddings original simply cannot. When Franklinrecorded the song, she was covering Redding.We could imagineas a thought experimentan artist today hearing only the Redding versionand recording a cover of it, but such a scenariois unlikely.29 Anyone performing or recording

  • Magnus, Magnus, and Mag Uidhir Judging Covers 367

    Respect now is covering the Aretha Franklinversion. This shift is captured best by thinking ofa cover as being of a canonical track, rather thanmerely indexing a cover to a song.

    Kania resists the idea that covers only occurpaired one-to-one with originals.30 We allow thatthe relation may not be one-to-one, but only be-cause the relation between a cover and canoni-cal versions can be more complicated. Becausepresent audiences know about both Dr. Dre andBen Folds, an a cappella performance of BitchesAint Shit might sensibly be called a cover of Dr.Dres original track or of Ben Foldss cover track.Yet it would be a mistake to conclude from thisthat covers are not connected to canonical ver-sions at all; it is better to conclude that a cover istypically connected to one canonical version butmay be connected to more than one. There is, inevery case, a musical-historical relation that re-lates the cover to a canonical version or versions.31

    Returning to the question of evaluation: Thereare two modes for evaluating transformative cov-ers, just as there are for rendition covers. The firstconsiders the cover in relation to the canonical ver-sion, and the second considers it regardless of thatrelation. In the first mode, contrasting Franklinsversion with Reddings, we find Franklins coverto be a deeper and more powerful trackevena deeper and more powerful songthan Red-dings original. In the second mode, consideringFranklins Respectwithout the explicit contrast,we evaluate it without mentioning Reddings ver-sion at all. Many people who hear the song onlyengage in the second mode of evaluation, a con-sequence of the fact that her track has becomecanonical.

    v. referential covers

    Consider the Meatmens cover of The SmithsHow Soon Is Now? This is clearly not a mimiccover, because there is a genre switch from 80salternative rock to punk.Moreover, there is a sub-tle change to the lyrics. Morrissey, lead singer ofThe Smiths, sings each chorus as I am human,and I need to be loved. Tesco Vee of the Meat-men echoes this in the first chorus, but changesit to I am inhuman, and I need to be fuckedand . . . I need to be killed in the second andthird choruses. Onemight initially think that theseare innocent changes to the lyrics. Just as Willie

    Nelson can sing about a girl from Austin, Texasand still be singing Paul Simons song Grace-land, one might say that the trio loved . . .fucked [and] killed is an interpretive substitu-tion. Yet the change here is importantly different.Graceland is about life tumbling in turmoil,and that is a diagnosis that might be offered bya girl in Texas as easily as by one in New York.How Soon Is Now? is a plaintive cry for the op-pressively shy, and being loved, fucked, or killedare radically different prescriptions for such a con-dition. So we must see the Meatmens cover as atleast transformative. There is still a further com-plication, however. The Smiths are (described un-charitably) a mopey band, and Morrissey (by hisown declaration) celibate.Whereas we are invitedto understand the I in The Smiths version of thesong asMorrissey or someone like him, the tone ofthe Meatmens cover does not invite us to under-stand it as being about Tesco Vee. Rather, the Iin the cover is most readily understood as mock-ingly about Morrissey himself. It is the mopey andasexualMorrisseywho the cover suggests needs tobe fucked, killed, or both. The Meatmens coveris not merely a distinct, derivative song. It is onewhich is partly about the canonical track and themanwho sings it. Its semantic content partly refersto The Smiths track.

    Call a cover of this kind a referential cover:one which instantiates a suitably derivative song(so counts as a cover by the condition from Sec-tion IV) but such that the new song is about theoriginal song or the canonical version.

    For referential covers, the secondmode of eval-uation collapses. The second mode of evaluationwould consider the Meatmens track on its own,rather than in contrast with the track that it cov-ers. In order to make this evaluation, however,we need to understand what the lyrics are about.If our interpretation is correct, then the lyrics inthe track are partly about the track that it cov-ers. So The Smiths version necessarily enters intothe evaluation. Failing to consider it would be abit like trying to understand Neil Youngs Ohiowithout knowing that Four dead in Ohio refersto the shootings at Kent State. The Meatmen singan original song titled Morrissey Must Die, andunderstanding it clearly requires knowing whoMorrissey is. Although the cover of How SoonIs Now? does not mention Morrissey or TheSmiths explicitly, it is similarly about them. So,for the Meatmens cover, any mode of evaluation

  • 368 The Journal of Aesthetics and Art Criticism

    ultimately requires considering the covers rela-tion to the canonical track. This will be true forany referential cover: any evaluation requires ref-erence to the canonical version.

    Consider also The Screamers 1970s cover ofSonny and Chers The Beat Goes On. It couldhardly be mistaken for Sonny and Chers so-very-1960s version. The music and lyrics are consider-ably refigured. The Screamers version is harsherand more confrontational. Sonny and Cher singthat Charleston was once the rage, but TheScreamers substitute Anarchy. Where Sonnyand Cher sing that Grandmas sit in chairs andreminisce, The Screamers sing, Pop stars sit inchairs and reminisce. Kids today are right to makea fist! And, after a line about cars becomingfaster, The Screamers add that Sounds are mov-ing faster, faster! . . . The song that The Scream-ers sing condemns slow, schmaltzy musicians likeSonny andCherit is about them. It is also a coverof a song by Sonny and Cher. So, we suggest, itqualifies as a referential cover.

    Mosser provides an extended discussion of SidVicious 1978 cover of My Way.32 The song waswritten for Frank Sinatra by Paul Anka in thelate 1960s, and Sinatras version is canonical. Inthe final verse, Sinatra claims to be a man true tohimself who say[s] the words he truly feels, butVicious inquires after a prat who wears hatsand cannot say the things he truly feels. So it isplausible to think that the considerably changedlyrics in Viciouss cover are in part a commenton Sinatra in the same way that the MeatmensHow Soon Is Now? is a comment on Morris-sey and The Screamers The Beat Goes On is acomment on Sonny and Cher. Properly evaluatingit requires keeping the original in view.

    Not all punk covers are referential covers.Shane MacGowans cover of My Way, likemany, reflects the Vicious cover musically butuses the lyrics from the Sinatra versionmakingit a rendition cover rather than a referentialcover. Punk bands like the Ramones and GreenDay have recorded numerous rendition covers oftracks that were not originally punk. In all, wesuspect that referential covers are fairly rare.

    vi. evaluation and ontology

    In the preceding sections, we distinguished mimiccovers, rendition covers, transformative covers,

    and referential covers. One may wonder whetherthis takes in all of the possibilities.

    Mosser distinguishes several different types ofcovers and arranges themon a continuumdepend-ing on how much the cover resembles the baseversion.33 Heputs reduplication covers (ourmimiccovers) at one end of the continuum because theyresemble the base version as much as possible andare the least original. He puts send ups (somethinglike our referential covers) at the other end of thecontinuum because they sound the least like thebase version. He places renditions somewhere inthe middle.

    We argue, however, that the types of coverscannot be sorted simply by how much the coverresembles the canonical track. First, a poorly ex-ecuted mimic cover might diverge dramaticallyfrom the canonical version. This does not sus-pend its status as a mimic cover. Quite the con-trary, the divergence is a defect just because ofthe evaluative criterion appropriate for mimiccovers. Second, we can imagine a referentialcover that sounds very much like the canonicaltrack. Suppose that the musical structure is pre-served, but that careful word substitutions aremade so as to subvert the entire meaning of thesong.

    More importantly, the categories that we haveidentified differentiate covers along two separateaxes. There are two questions to ask: Is the coveran instance of the same song? Does evaluationnecessarily involve reference to the canonical ver-sion? The possible answers create a two-by-twomatrix, each cell corresponding to one type ofcover:

    Evaluationrequiresconsideringthecanonicalversion.

    There is a mode ofevaluation thatdoes not requireconsidering thecanonical version.

    The coverinstantiates thesame song asthe canonicaltrack.

    Mimic cover Rendition cover

    The coverinstantiates adistinct, albeitderivative songfrom thecanonical track.

    Referentialcover

    Transformative cover

  • Magnus, Magnus, and Mag Uidhir Judging Covers 369

    Insofar as the two questions have clear, binaryanswers, the four types exhaust the possibilities.One might worry, however, that there are casesin which the questions do not have well-defined,yes-or-no answers. Perhaps you disagree with ourcategorization of one of the examples thatwe havediscussed and worry that there is no fact aboutthe covers themselves that will settle the disagree-ment. Here we reiterate that our primary concernis with evaluation rather than with metaphysics.There may well be covers that are not determi-nately of one kind rather than another. Even insuch cases, we suggest, attending to the relevantconsiderations facilitates evaluation. For example,to worry over whether The Screamers cover ofThe Beat Goes On is a transformative coverrather than a referential cover, one would needto consider the extent to which it is possible tounderstand it apart from the contrast with Sonnyand Chers track. To worry that it might just be arendition cover, one would need to argue that it isthe same song and so about the same thing that theoriginal was about (change and progress, perhaps)rather than a different song that subverts and is,in part, about the original. Even if these matterscannot be determinately resolved, these are re-warding reflections when the cover is a work ofart worth reflecting on.

    CRISTYN MAGNUS

    Albany, New York

    internet: [email protected]

    P.D. MAGNUS

    Department of PhilosophyUniversity at Albany, State University of New YorkAlbany, New York 12222

    internet: [email protected]

    CHRISTY MAGUIDHIR

    Department of PhilosophyUniversity of HoustonHouston, Texas 77004

    internet: [email protected]

    1. Each author contributed arguments and examples,and the ultimate form of the article was only possible be-cause of interaction between authors. They are listed in somesemblance of alphabetical order.

    2. cover, n. 1f, Oxford English Dictionary (Ox-ford University Press, 2013), http://www.oed.com/view/Entry/43347?rskey=TW9zLf&result=1#eid7893189(accessed May 14, 2013). The OED gives specimen usesof the phrase cover version going back to the 1960s. DonCusic writes similarly, The definition of a cover songis one that has been recorded before. Don Cusic, InDefense of Cover Songs, Popular Music and Society 28(2005): 171177, at p. 174.

    3. The 500 Greatest Songs of All Time, RollingStone, December 9, 2004, issue 963: http://www.rollingstone.com/music/lists/the-500-greatest-songs-of-all-time-20110407/patsy-cline-crazy-20110526 (accessed March18, 2013).

    4. Gabriel Solis, I Did It MyWay: Rock and the Logicof Covers, Popular Music and Society 33 (2010): 297318,at p. 301.

    5. Kurt Mosser, Cover Songs: Ambiguity, Mul-tivalence, Polysemy, Popular Musicology Online 2(2008): http://www.popular-musicology-online.com/issues/02/mosser.html (accessed March 18, 2013).

    6. Although the song is thinly specified, it is not sothin that anything could count as a performance or record-ing of the song. Franklin Bruno argues (and we agree) thatrock songs have enough content that they can be objectsof aesthetic evaluation over and above evaluation of tracksand performances; see Franklin Bruno, A Case for Song:Against an (Exclusively) Recording-Centered Ontology ofRock,The Journal ofAesthetics andArtCriticism 71 (2013):6574.

    7. See Andrew Kania, Making Tracks: The Ontologyof Rock Music, The Journal of Aesthetics and Art Criticism64 (2006): 401414.

    8. Although the event happens only once, perhaps theperformance qua object of aesthetic consideration can be re-peated. This complication does not matter for our purposeshere, but see Christy Mag Uidhir, Recordings as Perfor-mances, The British Journal of Aesthetics 47 (2007): 298314, and P.D. Magnus, Mag Uidhir on Performance, TheBritish Journal of Aesthetics 48 (2008): 338345.

    9. See Guy Rohrbaugh, Artworks as Historical Indi-viduals, European Journal of Philosophy 11 (2003): 177205, and P.D. Magnus, Historical Individuals Like Anasplatyrhynchos and Classical Gas, in Art and AbstractObjects, ed. Christy Mag Uidhir (Oxford University Press,2013), pp. 108124.

    10. Mosser, Cover Songs.11. The story was the inspiration for the 2001 filmRock

    Star, directed by Stephen Herek.12. The performance of amimic cover qua performance

    will probably sound somewhat different from the canonicaltrack. The performance is played in real time and the trackwas built up in the studio. However, this complication isnot especially about covers. It arises just as much when aband performs their own songswhich are better known fromrecordings.

    13. Kania, Making Tracks, pp. 401ff. We differ withKania on the related point of whether the object covered isthe song or the canonical track; see Section IV below.

    14. In later sections, we consider exampleswhere lyricalchanges do make for a different song.

    15. Story of the Song: Hazy Shade of Winter, TheBangles (1987), The Independent, November 26, 2010,

  • 370 The Journal of Aesthetics and Art Criticism

    http://www.independent.co.uk/arts-entertainment/music/features/story-of-the-song-hazy-shade-of-winter-the-bangles-1987--2144463.html (accessed March 18, 2013).

    16. YouTube hosts numerous examples; for example,http://www.youtube.com/watch?v=xzgiI5kJBe0 and http://www.youtube.com/watch?v=wVEl8AbMy5c (accessedMay 14, 2013).

    17. It would not be too pedantic to say that the groupsare covering Foldss cover rather than covering Dres trackdirectly. See Section IV below.

    18. Cusic, In Defense of Cover Songs, p. 174.Michael Rings focuses exclusively on covers of this kind.He calls them allusive covers: ones which make satu-rated allusions to the canonical version, making an in-direct reference to [the original] version simply by per-forming the song in question. Michael Rings, DoingIt Their Way: Rock Covers, Genre, and Appreciation,The Journal of Aesthetics and Art Criticism 71 (2013):5563, at p. 57.

    19. Mosser, Cover Songs.20. Rings writes, Knowing that her cover version

    stands to be heard by a wide variety of listeners for whoma given genre may signify a host of different things, the cov-ering artist may thereby choose to welcome a broad rangeof associations regarding the genres in play in her reset-ting. Rings, Doing It Their Way, p. 62. More generally,we should not suppose that an artist performing a coverhas anything deep or specific in mind. Perhaps he or sheperforms a cover merely because it is something that con-temporary artists do sometimes.

    21. Rings, Doing It Their Way, p. 60.22. Kania, Making Tracks, p. 411.23. One might say that the former evaluates Bs per-

    formance qua cover and the latter evaluates it qua songinstance. This risks slicing the ontology of events too finely,however, because both are evaluations of the renditioncover.

    24. They Did It Their Way. The Telegraphs Mu-sic Critics Select the 50 Best Cover Versions EverRecorded, The Telegraph, November 20, 2004, http://www.telegraph.co.uk/culture/music/rockandjazzmusic/3632028/They-did-it-their-way.html (accessed May 14, 2013).The 500 Greatest Songs of All Time, Rolling Stone:http://www.rollingstone.com/music/lists/the-500-greatest-songs-of-all-time-20110407/aretha-franklin-respect-20110516 (accessed March 18, 2013).

    25. Kania, Making Tracks, p. 410.26. They Did It Their Way, The Telegraph.27. Kania, Making Tracks, p. 411.28. Kania poses the paradox in this way: Imagine a se-

    ries of tracks A through Z, such that B is a cover of A, Cis a cover of B, and so on. Imagine further that C follows Bmostly in the respects it differs from A, D follows C mostlyin the respects in which it differs from B, and so on. So thereis nomusical or lyrical trace ofA left by the timewe reach Z.On his account, B covers the song instantiated by A ratherthan the track A, C covers the song instantiated by B, and soon. It follows (along with the transitivity of identity) that ZandA instantiate the same song despite sounding nothing atall alike. His reply is that the chain must break somewhere.Applying that answer here would throw him onto the firsthorn of the dilemma.

    29. Even if someone were to record a cover of Red-dings version for a tribute album, they would surely at leasthave heard Franklins.

    30. Kania, Making Tracks, p. 411.31. To be clear, we agree with much of Kanias account:

    that rock songs are thinly specified and that tracks are oftenthe proper object of aesthetic attention. We disagree justabout covers: on the question of whether covers are indexedto songs or to canonical versions and on the possibility ofwhat we call transformative covers.

    32. Mosser, Cover Songs, Section II.C.33. Mosser, Cover Songs.