Tangled up in Blue

The Shadow of Dylan and Stylistic Swerves in Early-Seventies Joni Mitchell

by Howard Wilde
Bloomsbury Academic
January 24, 2019

This is a chapter from Ruth Charnock's book "Joni Mitchell: New Critical Readings". See PDF download for full footnotes, diagrams, and attributions.

It has become a cliché of rock journalism to label Joni Mitchell a "female Dylan," something she has quite understandably resisted. In an interview with Morrissey in 1996, she noted that "They tend to lump me always with groups of women. I always thought: "they don't put Dylan with the Men of Rock; why do they do that to me with women?"1 Not only does Mitchell echo Morrissey's complaint about the phrase "female singer-songwriter" as a marked form, but she instinctively uses Dylan as her go-to example.2 In a lively polemic from 2012, Katherine Monk labels Dylan "the male Joni Mitchell," a counterblast against the patriarchy that underpins the construction of the rock canon (2012, 149).3 However, Mitchell's relationship with Dylan's work has always been double-edged: though the pair toured together and occasionally covered each other's songs, Mitchell later described herself as "almost anti-Dylan," and in a famous 2010 interview, suggested that Dylan's songs were "not authentic" and "not written honestly."4

This chapter seeks to explore Dylan and Mitchell's influences on each other, as well as Mitchell's stylistic influences on the next generation of songwriters, in a field characterized by male hegemony. I will try to avoid the traditional pitfalls of assessing Mitchell's work against Dylan's, as though his songwriting language were a definitive template by which hers should be judged. Likewise, I will not be aiming to produce a list of "echoes" between Dylan's work and Mitchell's, although of course there are many intertextual connections for those willing to search for them. Nor do I want to erase Dylan from a critical evaluation of Mitchell's songwriting influences. Instead, I want to propose the idea of Mitchell as "anti-Dylan": an equal and opposite aesthetic pole of influence.

The canonization of Dylan is an inescapable fact of history. He has been scrutinized by rock critics, thrust into the critical limelight whether he liked it or not, politicized as a "spokesman of his generation," compared favorably with Shakespeare, analyzed hagiographically by scholars like Wilfred Mellers, and projected as a successor to the English Romantics (Mellers 1984; Ricks 2003).5 Rightly or wrongly, he has come to stand metonymically for the very idea of songwriter as auteur, while Mitchell's work has been consigned to his penumbra. Even the most careful scholarly considerations of the Mitchell - Dylan relationship have tended to be slightly apologetic. Mitchell's work is more harmonically complex, arguably broader in its literary references, every bit as novel in its musical forms, and more wide-ranging in its range of musical influences. Yet so many commentators, including a sample I will critique later in this chapter, insist on "validating" Mitchell's work in the context of Dylan's, as though the mere comparison were the greatest compliment they could bestow. This critical imbalance can legitimately be critiqued along gendered lines, not least because the corpus of rock journalism is so overwhelmingly male-dominated.

In this chapter, I will aim to redress this balance by identifying some of the defining aesthetic criteria by which critics have traditionally validated Dylan's work, and suggesting an alternative, largely antithetical, set of criteria which apply to Mitchell. I will not be attempting a sociological critique of rock journalism and its attendant biases: there are already plenty of excellent critiques in the musicological literature (e.g., Leonard 2007; Hopper 2015). Instead, I will propose a binary division between Dylan and Mitchell's songwriting styles based on close readings of selected songs. I will propose that Dylan's work is essentially sermonic: his songs can be apocalyptic, vatic, oblique, accusatory or confessional, and their narrative complexity (his main claim to auteur status) is offset by harmonic simplicity. Mitchell's work, by contrast, is often conversational, intimate, questioning, uncertain; her complex narratives are often enriched by some equally complex and ambiguous harmonies, influenced by jazz, classic-romantic music, painting, and high modernism. Instead of "validating" Mitchell's work by occasional "nods to Bob," I will suggest that she achieved a decisive stylistic swerve on the album Blue, which in turn exerted a recursive influence on Dylan's output, as well as providing a blueprint for a new generation of songwriters.

Canonization and influence as anxiety

In considering the ambivalent relationship, both personal and aesthetic, between Mitchell and Dylan, it is worth considering two theoretical topics that are broadly intertwined, and generally more developed in literary criticism than in pop-rock musicology: first, theories of the canon; and second, theories of influence. Both are ideologically problematic and politically polarized, and although much of the debate will be outside the scope of this chapter, a number of key concepts might be useful. Central to these ideas are the writings of Harold Bloom, a figure who has divided critics with his unfashionable defense of the "high art" canon, his use of the terms "strong" and "weak" poets, and his bold use of the word "authenticity" without the layers of anxious irony that accompany that word in twenty-first-century literary and musicological circles.6 More controversial still is his contempt for what he has termed the "school of resentment" (postcolonial theory, feminism, black literature) (Bloom 1995, 4). In particular, Bloom is notoriously dismissive of the idea that popular songwriting might be poetry, let alone "great" or canonical poetry. In an appendix, Bloom updated his canonical list of twentieth and twenty- first-century poets to include John Ashbery, Geoffrey Hill, Elizabeth Bishop, Marianne Moore, and others (1995, Appendices A - D). Although it is not an especially sexist list, it is notable for its exclusion of musicians, or indeed anything "populist": Dylan, Cohen, Mitchell, and others are notably absent. This position echoes the "high art" purism of the Marxist critic Theodor Adorno, who regarded the "serious" protest songs of the sixties (by Joan Baez and others) as fatally tainted by commodification and banality.7 Slam poetry, said Bloom, was "the death of art" (in Somers-Willett 2009, 21).

Other poetry scholars, however, have been passionate advocates for the canonization of Dylan in particular. Christopher Ricks's ambitious Dylan's Visions of Sin aims to locate Dylan's body of work within the tradition of nineteenth-century Romanticism, offering cross-readings between Dylan songs and poems by John Keats and Alfred Tennyson.8 This has the effect of reinforcing Dylan's canonical status by removing him from his immediate milieu, and giving him the aura of the misunderstood outsider: "the poet and the painter, far behind his rattled time" (as Dylan himself put it).9 In this text- centered critical approach to Dylan, we are invited to "read" his "work" (both words are loaded) outside the traditions of rock and roll, beat poetry, the Never Ending Tour, the blues, and all the messy, rhizomatic, multicultural cross- threads that make Dylan's songwriting what it is.10 If this approach elevates one aspect of Dylan's writing at the expense of others (the textual over the performative), it seems even more problematic when dealing with the melodic and harmonic complexities of an artist like Mitchell.

Dylan himself has always been ambivalent about his canonization by the critics. In fact, critic-mockery has been a common theme of his writings, interviews, and songs. It is probably fair to say that he has written as many songs about his critical reception as about war and politics (the two topics often ascribed to him by lazy critics): "Ballad of a Thin Man" (1965) satirizes an earnest Bob-worshipping journalist; "It Ain't Me, Babe" (1964) is often delivered in a finger-pointing philippic against his adoring audiences; "Maggie's Farm" (1965) captures the anger of a folksinger dragged unwillingly on to a canonical pedestal: "They say 'Sing while you slave' and I just get bored." Later, however, in the Chronicles, Dylan is at pains to rebrand himself a folksinger once more, casting a snide barb in the direction of the melancholic poet Archibald MacLeish (Dylan 2004, 141). He also suggested that he had deliberately released light country songs on Nashville Skyline (1969) to put critics off the scent. So, Dylan's entire career has been infused with a kind of meta-critical jouissance, flirting with canonical immortality but never quite letting the critics have the upper hand.

While the idea of a canon is still common currency in journalistic circles (now more than ever, in the age of top-hundred lists and clickbait), it has been largely discredited in the musicological literature.11 Marcia Citron's deconstruction of patriarchal narratives in the (classical) musical canon was followed by Cerys Wyn Jones's feminist critique of the rock canon, and a study by Ralf von Appen and André Doehring, which hinted that the idea of a rock canon was a preserve of middle-aged white men, subconsciously asserting their own hegemony (Citron 2000; von Appen and Doehring 2006; Jones 2008). Indeed, von Appen and Doehring's meta-canonic list (compiled from dozens of journalistic top-hundred lists), is striking for its exclusion of female, LGBTQ, and nonwhite artists. This bias has been reflected every year in my undergraduate classes; when students have been asked to list the ten "most influential albums" released since 1960, fewer than 5 percent of their nominations feature a female artist, let alone a female songwriter.12 The annual (and now painfully predictable) omission of Joni Mitchell seems significant, given her vast critical acclaim and the diverse ambit of her influence. Why is Joni Mitchell's work so little-known, and so under-appreciated, among popular music students? (Dylan's name features in their lists, although they often admit to unfamiliarity with his work.)

Of course, my questionnaire uses loaded questions: Why, for instance, do we tend to privilege albums over singles? Why, almost 30 years after Christopher Small's radical book Musicking: The Meanings of Performing and Listening, do we still try to reify the rock album as an artefact?13 And what is it to be "influential"? To this end, I want to invoke Bloom's most celebrated idea: the idea of influence as anxiety. Poets, argues Bloom, invoke other poets, and cannot become poets without first having been inspired by other poets. In an ironic echo of the leftist intertextual criticism largely despised by Bloom, poems (for him) achieve their meaning only in relation to other poems. As we have already seen, Bloom has shown no interest in extending his theory to popular music, but the idea of influence as anxiety might be helpful in explaining changes of style, especially in key albums like Joni Mitchell's Blue: a radical departure from the Dylanesque norm of folk rock at a critical historical juncture. In particular, Bloom's theory of influence invokes the idea of the "swerve": a break with tradition that comes from the anxiety of sounding too much like one's precursor.14

In fact, it might be fair to say that, of the interesting singer-songwriters to emerge around the revolutionary year 1968, no one really sounds less like Dylan than Joni Mitchell. Unlike him, she did not extensively borrow the twelve-bar blues form; she did not learn traditional guitar tunings by default but reinvented tunings according to her own caprice; she was largely an autodidact when it came to piano and guitar technique (like her fellow Canadian, Leonard Cohen); she was a poet and painter before she was a folk singer; she might have adopted the folk-rock idioms of the time for Clouds but certainly did not style herself as a folk troubadour in the mold of Baez; she distanced herself from certain political movements and causes, preferring environmentalism to radical feminism, and so on. Her status as a Canadian woman might also have lent her the aura of an outsider, perhaps making it easier, rather than more difficult, to escape the oppressive shadow of Dylan and the inevitable comparisons. In Bloom's model, the act of "swerving" often takes the form of invoking an earlier, or "other" style in order to escape the burden of the immediate past: thus Stravinsky would invoke Machaut, Gesualdo and jazz as a way of wriggling free from Russian Romanticism; the Second Viennese School would evoke Bach not as a kind of sentimental archaism but as a means of renewing the musical lexis after two generations of Austro-German Romantic tradition.15 So it is tempting to read Mitchell's excursions into jazz (on Hejira) and progressive tonality (on Blue) as "swerves," conscious or otherwise, from a hegemonic tradition: in this case, that of the three-chord folk song as exemplified by, say, Joan Baez's covers of Dylan. Inevitably, critical reviews of Mitchell's work in the seventies would invoke Dylan. When they did, however, it was not as a stick with which to beat Mitchell, but more often a way of sneering at His Bobness, knocking him off the pedestal they had constructed. After all, by the seventies he was all but buried by the critics: they were already searching for a new Bob Dylan, and the lackluster Self Portrait, released the year before Blue, was widely regarded as the end of Dylan's influential phase, and the beginning of a period of mediocrity. David Bowie's "Song for Dylan" (1971) and Wainwright's later "Talking New Bob Dylan" (1991) both hint at the anxiety of the songwriter in a post-Dylan era.

The tone of this review of The Hissing of Summer Lawns, by Stephen Holden for Rolling Stone in 1976, is not untypical:

[Mitchell] has amassed the most impressive body of work of any post-Dylan singer-songwriter, elaborating the free-form narrative ballad form that Dylan thrust into popular song, and polishing a melismatic melodic line more flexible than Dylan's and many times more sophisticated.16

Better Than Dylan, then, but only to be read and appreciated through the prism of Dylan, and only better in respect of her vocal embellishments, not necessarily her imagery or narratives. It is also debatable whether Mitchell's songs are really reducible to "narrative ballads," at least in the folk sense of that term, with all the moralizing and storytelling that we find in Dylan.

More recently, in his encyclopedic study of Mitchell's work, Lloyd Whitesell devotes a section to the influence of Dylan (his excellent musical analyses propose a typography of her song forms, again tellingly using Dylan songs as a template). However, Mitchell herself refers very sparingly to Dylan songs when discussing her influences. One exception occurs in the following interview, cited more elaborately by Whitesell: "When I heard "Positively Fourth Street," I realized that this was a whole new ballgame: now you could make your songs literature" (2008, 77). In a later remark about the same song, she praised the "direct, confronting speech, commingled with imagery." This remark, for Whitesell, constitutes a "galvanic spark" originating from her "first encounter with the seminal Dylan song." He calls this moment "an early conversion experience":

We know from Mitchell's subsequent career that this early conversion experience did not cancel out her love of rock 'n' roll. The Dylanesque model of serious poetic ambition merely took its place alongside the Little Richard model, in an expanded understanding of what words in pop songs can accomplish. (Whitesell 2008, 77)

This account is not without its problems, though. For one thing, it assumes a dichotomy between the "high art" ambitions of Dylan as auteur and the populist aesthetic of Little Richard; yet Little Richard was an enormous early influence on Dylan, too.17 Many of Dylan's songs, not just the simple Elvis pastiches on Nashville Skyline or The Basement Tapes, draw on rock 'n' roll to a much greater extent than any Mitchell song; the country blues is a staple of Dylan's songwriting, from "Highway 51" to "Thunder on the Mountain" and all points in between. Equally, the choice of "Positively Fourth Street" is interesting: it was never released on an album, and so stands somewhat outside the canonical "body of work"; it is hardly "literary" in its language; it contains no obvious poetic imagery, metaphor or poetic conceits. It is certainly direct and vituperative, more so than any typical pop song of the time; it is also uncompromisingly strophic, without any hook, chorus, bridge, or other conventions of current chart pop. But literature? We might be skeptical about taking her remarks at face value here, given the breadth of literary allusions that pervade her own work. In any case, she was already immersed in Western art songs, referencing German lieder and jazz. She needed no Damascene conversion to song-as-literature by a (reluctantly) canonized Dylan.

That said, it is impossible to escape the fact that Mitchell's career began in an era of unbridled Dylanolatry. Singer-songwriters needed to take a position vis-à-vis Dylan, whether imitating him passively, covering his songs, responding to his words in what would nowadays be called a "backatcha," paying homage, or self-consciously swerving from his influence. Several singers began their careers by covering Dylan songs in a respectful fashion, often smoothing down their rough edges: Judy Collins, Joan Baez and Nico among others. Others executed more obvious swerves and symbolic rejections of Dylan's influence: John Lennon's symbolic rejection of Dylanolatry came in his famous post-Beatles song "God" ("I don't believe in Zimmerman" - emphasis mine) (1970), while Paul Simon's "A Simple Desultory Philippic" (1966) was a perfect satirical impression of Dylan that captured the absurdity of the Bob-worshipping zeitgeist. The anxiety of influence was quite palpable. If Joni Mitchell had stopped recording after Ladies of the Canyon, her career might legitimately have been characterized as "post-Dylan," along with his other imitators. But the harmonic and prosodic adventures on Blue evoke not a post-Dylan landscape but a vigorously anti-Dylan language, whether conscious or otherwise.

Song as sermon; song as conversation

In the following discussion, then, I would like to argue the case for Joni Mitchell as an opposite pole of influence, at least equal to Dylan but diametrically opposed in several ways: linguistic register, harmonic vocabulary, phrasing, diction, rhyme-schemes, and musical structure. To summarize this dichotomy, we might think of two broad tropes of folk-rock songwriting: song as sermon and song as conversation (see Table 7.1).

Of course, not all Dylan songs are literal sermons, and not all Mitchell songs are conversations (although "Hard Rain" and "A Case of You" might be useful exemplars of each). I have also hesitated to map this loose schema on to a gender binary, because plenty of female songwriters use a sermonic style (Ani DiFranco comes to mind), while some male writers aim for a prose register (such as Maynard James Keenan, who has often acknowledged Mitchell's influence). It would also be too simplistic to regard the division as "political versus apolitical": Mitchell's songs are often politically charged without lecturing, as when she sings about the Magdalene Laundries. Equally, the caricature of Dylan as a political songwriter fails to do justice to his hundreds of love songs.18

The musicologist Dai Griffiths hints at a similar kind of dichotomy between the self-consciously poetic and the conversational. In his interesting concept of the "anti-lyric," the term "lyric" is double-edged: in its literal sense it denotes the words of a popular song, but it also carries historical connotations of lyric poetry (as distinct from epic poetry): poetry which foregrounds the writer's emotions and situates the self at the center of the narrative (Griffiths 2009). Griffiths defines the "lyric" tendency as "words that tend to be like poetry" and "anti-lyric" as "words that tend to be like prose" (2009, 42). A famous example occurs in "A Case of You":

Just before our love got lost you said
"I am as constant as the northern star"
and I said, "Constantly in the darkness
Where's that at?
If you want me I'll be in the bar"

The structural rhyme (star/bar) is partially erased by the "and I said," shifting the rhyming word into the middle of the line, and moving the banal, conversational signifiers (you said, I said) to the end of the musical phrase. In the narrative, Mitchell's male interlocutor comes out with a vatic Julius Caesar quotation and is gently cut down to size by the narrator, switching effortlessly to world-weary vernacular. This is just one example of the kind of ironic bathos that characterizes her writing on Blue.

This mode of delivery is in marked contrast with Dylan's typical style of declamation. Dylan's rhymes can be astonishingly playful and creative, but they almost always have the effect of reinforcing the sense of teleology and closed structure.19 Indeed, some of Dylan's most impressive comedic rhymes appear in songs with the most pointedly moralistic message. A famous example is "Hurricane" (1976), where internal rhymes jostle with structural end-rhymes in a way that prefigures hip-hop, not least in anger and visceral energy:

And ride a horse along a trail
But then they took him to the jail / house
Where they tried to turn a man into a mouse.

This is no random, improvised piece of rhyming virtuosity, though: it is part of the song's design, entirely consistent across the song's eleven verses. The effect of the short penultimate line is teleological, and essentially sermonic: to drive toward the angry, righteous peroration (in this example, the word "mouse" is prolonged passionately for a whole bar, in a way that simply would not be possible on the page, or in a poetry reading). A similar effect is achieved in "Tangled up in Blue" (1975), in which the tagline is prepared by a shorter line, in order to drive to the final cadence on "blue": this word is progressively stretched from verse to verse, in both the album version and several live recordings. This drive toward the peroration is starkly at odds with Mitchell's style of delivery in many of the songs on Blue. This example, from "California" is typical: "They won't give peace a chance / That was just a dream some of us had." The syllable "had" (rhyming with "bad") falls on a weak beat, giving a sense of bathos and understated resignation. If Dylan's rhymes take us by surprise us when we hear them, Mitchell's rhymes are easy to miss, even on the page.

Another distinctive feature of Dylan's declamation is the very deliberate variation in syllabic density - to use a term of Leonard Cohen's (Zollo 1997, 341). This is a parameter of song which does not normally exist in written poetry: the relationship between the syllable-count of the line and the musical meter. Some Dylan songs have varied syllabic densities which make them appear unconvincing on the page, but extremely powerful in their sung delivery, as here, in "Not Dark Yet" (1997), where eighteen syllables are crammed into a two-bar phrase:

Every nerve in my body is so vacant and numb
I can't even remember what I was I came here to get away from.20

Again, the effect is teleological: the tone may be conversational, but the narrative is still essentially a sermon, prophesying the end of the world. Indeed, Dylan's embrace of the "song as sermon" aesthetic is a defining feature of his style, from the early "A Hard Rain's A-Gonna Fall" (1964) (one of his very rare unrhymed songs), through his Christian period ("Gotta Serve Somebody" from 1979) to the various apocalypse songs, some of which (from 1963's "Talkin' World War III Blues" to "Tempest," 2012) are faintly comedic. The binary opposite of the "preaching" song is the confessional song, something occasionally embraced by Dylan, when he moves from the pulpit to the confession box ("Every Grain of Sand" from 1981, for example).

Mitchell is famously averse to the idea of the "confessional" songwriter. In a recent article for the Guardian, Alexandra Pollard has argued that the label is almost always applied to women poets (Sylvia Plath, Ann Sexton, and others), and quotes this particularly offensive (but not untypical) comment on Plath's work by a male reviewer: "The personal character of the confessional detail is embarrassing, and the tone of hysterical melodrama which pervades most of the writing is finally irritating."21 Mitchell recently expanded on this idea, suggesting that there are two types of confession: the involuntary confession of sins under torture, or the "voluntary" self-abasement of the confession box (Mitchell, cited in Pollard 2015). Unlike so many of Dylan's songs, Mitchell's are never sermons, lectures, or confessions.

"The Last Time I Saw Richard": songwriting and cubism

"The Last Time I Saw Richard," the last track on Blue, is neither moralistic nor confessional. Instead, it is an early example of the conversational trope, in which a female narrator, world-weary and slightly cynical, looks back on a conversation in a bar with a male interlocutor.22 Such a simple narrative device is rare in the pop songs of the late sixties and early seventies, but it occurs in later examples by Alanis Morissette, like the epistolary song "Unsent" (1998) and "Joining You" (1998) in which the narrator writes supportively to a friend who was contemplating suicide. In its narrativity, "Richard" frames the conversation in the past tense, shot through with nostalgia: to that extent, it harks back to the nostalgic song cycles of Schubert and Schumann (both Winterreise and Dichterliebe begin with memories, and the narrative arc of both cycles is clearly framed in a happier past). The device has already been prefigured in the preceding song "A Case of You," in that poignant line "Just before our love got lost, you said," but here it becomes a structural framing device: the conversation is interspersed with factual banalities of suburban life (the dishwasher and coffee percolator being the early-seventies equivalents of the "gramophone" famously evoked in Eliot's The Waste Land).23 I mention Eliot here, not as a way of "legitimizing" Mitchell's work through comparisons with dead male poets (as Ricks does in his Dylan - Keats cross-readings), but to show just how vastly Mitchell's idiolect differs from that of Dylan. If Dylan has one foot in the nineteenth century, far behind his rattled time, Mitchell's literary reference points are firmly rooted in the twentieth. If Dylan's images resonate with Romantic poetry and antebellum Americana, Mitchell's are alive with modernism.24 The gently prosaic references to dishwashers and coffee percolators in "Richard" would seem very out of place in a Dylan song.

The musical structure of "Richard" is similarly conversational. It feels largely improvised, although a careful comparison of the three verses reveals a very deliberate strophic structure. Likewise, there are variations in syllabic density, perhaps even more dramatic than those in Dylan: the "gorgeous wings" line is set to an almost madrigal-like melisma, but its sentimentality is undercut by the prosodic diction of the verses, sung to a speech-like oscillation of fifth and third scale degrees:

The last time I saw Richard was Detroit in 68 and he told me All romantics meet the same fate, someday. (Emphasis mine)

This is the diametrical opposite of Dylan's style of declamation: instead of the teleological drive toward the rhymed word at the end of each line, the delivery effaces the rhyme, and shunts it into the middle of the line. It is not quite the same as the internal rhyme mentioned by Whitesell in his reading of "I Had a King": in that earlier song, each line contains an end-rhyme and a concealed rhyme as a deliberate structural device, but here there seems to be an equally deliberate dovetailing between the closures implied by the rhymes and those articulated at musical phrase endings (2008, 17).

However, it is in her musical structures that Mitchell takes the most decisive stylistic swerves from the folk-rock tradition and forges a genuinely new technique that was to prove influential for songwriters of the next generation. Several commentators have found echoes of jazz in this early phase of her career (e.g., Whitesell 2008, 22 - 23 on Court and Spark),25 and it is possible to hear "Richard" as a bridge between the folk idioms of her earlier writing and the rich additive harmony and quasi-improvisatory storytelling of her seminal jazz album, Hejira. Though Dylan's harmonic language is rooted in the blues, he almost never uses the additive harmonies of jazz, and his tonal centers are nearly always unambiguous. Conversely, "Richard" is full of tonal ambiguity, reflecting the emotional ambivalence of the lyric.

The identification of a tonic is problematic: many online transcriptions, including some endorsed by Mitchell herself, notate the song in two sharps, perhaps influenced by the prolonged D chords of the intro, and perhaps by the cryptic final chord of Bm11 (a "jazz" chord or Stravinskian "bitonal" chord, depending on how we want to interpret it). However, the body of the song is clearly in G major, even though this is undermined by the ambiguity of the surrounding piano episodes. Whitesell, in his taxonomy of tonality and modality in Mitchell's songs, assigns "Richard" to the "polytonal" category, as distinct from the commoner "polymodal" category, which implies chromatic variations in melody lines over an undisputed tonic (Whitesell 2008, 122). However, "polytonal" does not imply two simultaneous tonics, as in sections of Stravinsky's Petrouchka or the bitonal folk-melody arrangements in Bartók's Mikrokosmos. Instead, it implies tonal ambiguity: there are no perfect (or, indeed plagal) cadences to provide closure in the key of G major (the only V- I progressions being at the start of each strophe), and the outer sections undermine the tonal center.

Tonal ambiguity can be explained in two very different ways: first, as a modernistic tendency paralleling the "extended tonality" found in the art music of the late nineteenth century and codified by theorists such as Deborah Stein in her study of the songs of Hugo Wolf (Stein 1985). In this model, the songwriter uses tonal ambiguity to "swerve" from an established framework of tonal closure: it is essentially an avant-gardist gesture. The second explanation is that tonal ambiguity arises in rock songs because our Western-centered paradigm of tonality is biased towards a single, controlling tonic, and in trying to interpret a particular blues or rock sequence in terms of a single key center, we are misreading one musical language through the prism of another.

A famous example of this idea is in Lynyrd Skynyrd's "Sweet Home Alabama" (1974): the familiar D-C9-G loop can be parsed as I-bVII-IV in D mixolydian or as V-IV-I in the key of G. Philip Tagg, in his excellent taxonomy of loops and shuttles in pop songs, effectively cuts the Gordian knot by suggesting that such loops are merely motions from one point to another, without any need to decide which is the hegemonic tonal center (Tagg 2014, 371 - 400). In other words, by asking whether G or D is the "real" tonic, we may be asking an unnecessarily loaded question.

In "Richard," however, the tonal ambiguity seems both deliberate and poetic: the three sung verses can be explained quite unproblematically in terms of a G major tonic (admittedly, not with a "common practice" type of voice-leading or harmonic progression), while the piano sections flirt with D major, F sharp minor and B minor.26 If we disregard the piano introduction and outro, the three strophes all outline a "flat side" chordal loop of [G - Am (x2)] - C - F - G, corresponding to I - ii - IV - bVII - I in G mixolydian.

Behind the additive dissonances of the surface, then, the schema is not unlike that of Cohen's "One of Us Cannot Be Wrong" (1967), which also uses a I - ii oscillation, and moves to the chord of bVII at the end of each strophe. While the internal structure of the strophes emphasizes the flat side of the circle of fifths, each strophe is linked by a more traditional circle-of-fifths progression from the sharp side: A - D - G (or V/V - V - I). These underlying progressions have plenty of precedents in the folk rock of the late sixties and early seventies, but here the vocal line takes a startling swerve into a jazz aesthetic that prefigures Hejira: in the line "look at your eyes, they're full of moon," Mitchell sings an unprepared ninth above the bass. Likewise, in the sigh-like codettas of each strophe ("pretty lies," "love so sweet" and "dark café days"), the vocal melody presents unresolved elevenths over the A in the bass. The last two notes of the melody line in the song ("café days"), and indeed the entire album, would be consonant with the tonic, but the tonic has yet to arrive. When it does, it is a brief closure, eclipsed by the cryptic Bm11 that closes the track: a dissonant collision of two chords (Bm and A) heard simultaneously.

This kind of ambitious harmonic language is the antithesis of the Dylan folk-rock aesthetic. Although Dylan talked about the flattening out of time in his lyrics (specifically "Tangled up in Blue"), there is nothing in his work that comes close to this kind of structural stasis in the music, where tonal centers occur synchronically.27 To find an origin of this startling style-shift, we might look instead to high-art modernism. In a famous book on Stravinsky, Roman Vlad introduces the intriguing concept of musical cubism. Cubism in painting involves the simultaneous representation of objects from different perspectives, yet the inherently non-representational nature of music means that "cubist music" is perhaps, at best, a metaphor.28 Vlad, however, points to a number of interesting chords and heterophonic details in Stravinsky's music which disrupt the tonal syntax by presenting tonics and dominants simultaneously: an example is the dissonant chord that closes the Symphony in C (1940). Edward T. Cone, in an early analysis of the Symphonies of Wind Instruments, offered a similar reading in which familiar chord progressions were made strange and unsettling by being interspliced in a kaleidoscopic fashion (Cone 1968, 18 - 26). It is tempting, then, to suggest that Mitchell's songwriting language draws as much on her experience as a painter as on her exposure to the folk-rock tradition. Blue is, after all, a "painterly" album, suffused with references to colors: "I miss my clean white linen"; "Let's have a round for the bright red devil," "I am a lonely painter / I live in a box of paints," as well as "Little Green" and of course "Blue" itself, with its references to ink and tattoos.

As a poet and a painter before she became known as a songwriter, Mitchell was already equipped with a songwriting paint box that allowed her to eschew the Dylan paradigm with greater ease and assurance than many of her contemporaries. In fact, her influence is keenly felt on Dylan's Blood on the Tracks, released four years after Blue. It was a time when Mitchell's star was in the ascendant, at least in terms of critical reception, rather more than his own. "Tangled up in Blue" may not be a direct reference to the Mitchell album, although one critic recently claimed that Dylan told him precisely this.29 However, it was written in a renewed creative phase in 1974 - 1975, after a brief hiatus filled with the study of painting and modernist poetry. He himself suggested that the song was concerned with the flattening of time, and the use of discontinuous, fragmented narrative. It is certainly the first, and perhaps the only, Dylan song to use the quasi-cubist, nonlinear narrative more familiar from Hollywood flashbacks or the modernist novel. Though its melodic and harmonic structure is infinitely simpler than "Richard" or indeed anything on Blue, its startling, almost filmic imagery ("rain falling on my shoes") and its seamy adult themes of drug dealers in basements seem to suggest a stylistic affinity with the Mitchell of "acid, booze and ass / needles, guns and grass."30

Influence and legacy

The idea of influence as anxiety, then, seems useful to the extent that it can help us identify decisive stylistic swerves, not just in comparing one songwriter's style with another, but in tracing the stylistic evolution of each. The anxiety of influence might apply just as well to the burden of living up to one's own stylistic reputation: so Dylan's career is littered with anti-Dylan swerves, in which he ritually disavows earlier styles, and challenges the caricatures erected by the critics. The most dramatic of these swerves came in his Christian period, beginning with Slow Train Coming in 1979, but his touring career has been marked by dramatic changes in the interpretations of his own songs. Likewise, Mitchell's long recording career is marked by significant style-shifts, the tonal ambiguities and complexities on Court and Spark (1973) and Hejira (1976) being prefigured by the assured, relaxed harmonic complexity of "The Last Time I Saw Richard."

In turn, Mitchell herself casts a long shadow, and provides a source of inspiration (and perhaps anxiety) for a later generation. Inevitably, Mitchell comparisons have been foisted on female songwriters: Laura Marling, for instance, was being dubbed the successor to Joni Mitchell at a very early stage in her career, and her excellent "Reversal of the Muse" series explores the role of women in the creative arts (presenting women as inspired creators rather than as muses).31 Alanis Morissette, for example, spent some time in India between her first and second albums, wrestling not so much with the "anxiety of influence" as the anxiety of her own fame. If the songs on Jagged Little Pill (1995) were Dylanesque and sermonic (even down to the harmonica solos and the political anger on the closing track, directed at the misogyny of the Catholic Church), the more striking tracks on Supposed Former Infatuation Junkie (1998) seem unthinkable without Mitchell: in particular, the radical tonal scheme of "Joining You," which alternates conversational verses (in a static C minor) with sermonic choruses (modulating sharpwise through C, G, D and A).

But is there a danger, in reassessing Mitchell's legacy, of casting her as a role model purely for women songwriters? As David Bennun remarks: "When Mitchell's influence is mentioned, it is almost exclusively in terms of how she inspired those female singer-songwriters, as if a female artist may be the creative begetter only of other women."32 Perhaps, although both Marling and Morissette have embraced the "Bob" trope in their work every bit as much as the "Joni" trope, while a growing number of male singers and songwriters cite Mitchell as an influence. The sixties phenomenon of women covering Dylan songs has given way, encouragingly, to a growing trend of men covering Mitchell songs: Ian Shaw, James Blake, and Rufus Wainwright among the more creative examples.

I have not tried to claim that Mitchell is superior to Dylan (although her musical language is undeniably more complex), but that her body of work has shown a generation of songwriters that a creative auteur does not need to tread in Dylan's footsteps. Just as the tradition of "progressive" songwriting from Sgt Pepper to Pink Floyd would be unthinkable without Dylan, so a growing body of music would be unthinkable without the radical stylistic excursions of Blue. Rather than interpreting Mitchell's work as an antithetical "reaction" to Dylan (as per Bloom's lines of thought, and the critical positions that regard her as "post-Dylan") it might be more fruitful, and fairer to both, to regard them as equal and opposite poles of influence: two artistic traditions which bifurcated significantly in the early seventies. Whether Dylan's influence was a source of creative anxiety in any literal or psychological sense is a moot point: as I hope to have shown, the internal evidence of her songs seems not to suggest some sort of Freudian wrestling with a precursor, but rather a startlingly fresh idiolect that showed a generation of songwriters that there is another, at least equally interesting, way of approaching the craft.


Printed from the official Joni Mitchell website. Permanent link: https://jonimitchell.com/library/view.cfm?id=4658

Copyright protected material on this website is used in accordance with 'Fair Use', for the purpose of study, review or critical analysis, and will be removed at the request of the copyright owner(s). Please read 'Notice and Procedure for Making Claims of Copyright Infringement' at JoniMitchell.com/legal.cfm