A resisting performance of an Appalachian traditional murder ballad: giving voice to "Pretty Polly".
Dellie Norton, traditional ballad singer, Madison County NC, as told to the author by Sheila Kay Adams, Norton's great-niece, 7 November 2004
SCENARIO 1: MY PARTNER'S MOTHER INTRODUCES me to her friends: a white, upper-middle-class, heterosexual couple. They are in their late sixties or early seventies and are from northeastern Ohio, just one county away from those considered Appalachian. Clearly a part of the symphony set, their faces brighten when they learn I'm a musicologist stopping off in Ohio on my way to a music workshop in North Carolina.
"What are you going to do?"
"I'm going to study old-time music, specifically, the banjo."
"What is that?" he says with a raised eyebrow.
"Old-time music," I say very simplistically, since I can tell they are already losing interest, "is the kind of music that predates bluegrass and contemporary country music. It's a kind of early country music, actually."
"Oh," he says with a look of amused condescension, "I didn't think you could call that music."
Scenario 2: A friend who loves old-time music and bluegrass and who plays fiddle, guitar, mandolin, and Dobro decides that she can't listen to most old-time and bluegrass bands anymore since so many of the songs are misogynist. It is the violent ones that anger her the most, especially the murder ballads, in which the woman is horrifically killed and her murderer suffers few consequences. The Country Gentlemen's recording of "River Bottom" is particularly offensive to her: "with a clothesline tied around her knees ... I'm so glad to put an end to that disease."
These true stories illustrate a dilemma for the feminist lover of old-time, bluegrass, and Appalachian music. (1) The first scenario demonstrates a widely held position: Appalachian music, with its unabashed banjo twanginess, its hoedown fiddle, and its high, strained, and dispassionate vocal style, is a vivid aural representation of Appalachia itself and as such cannot be considered music worthy of serious study, much less enjoyment. As Ronald D. Eller writes, "Appalachia may likely have replaced the benighted South as the nation's most maligned region.... Always part of the mythical South, Appalachia continues to languish backstage in the American drama, still dressed, in the popular mind at least, in the garments of backwardness, violence, poverty, and hopelessness once associated with the South as a whole. No other region of the United States today plays the role of the 'other America' quite so persistently as Appalachia." (2)
Of course, there is an audience that enjoys escaping into precisely this "other America," with its somewhat exotic rural sounds and modal melodies that can conjure up images of a simpler and more authentic era and place, far from urban noise and confusion. (3) Some of these listeners may choose to ignore the misogynist lyrics or perhaps not even notice them, while others may explain them away as artifacts of past cultural norms. (4) Others, however, like the woman in the second scenario, may feel conflicted, eventually choosing to abandon a favorite genre of music that is both "maligned" by "the popular mind" and unmistakably misogynist for the listener, fan or not, who attends to the lyrics of much of its repertoire. I would like to examine another option for the woman in the second scenario. Perhaps it is possible to find or create performances of Appalachian music that retain the exuberance of the fiddle, the twang of the banjo, or the detachment of the voice (performances that, in effect, sound Appalachian) while presenting the violent lyrics in a way that does not celebrate or ignore their misogyny. The most challenging songs will, of course, be the murder ballads, which pervade American folk music and are central to the southern Appalachian repertoire. (5)
The plot of these "murdered sweetheart" ballads follows a well-established formula: a young woman is lured away from home by her lover to a secluded spot on the pretext of marriage or discussing marriage; presumably, she is pregnant. Once they go away together he kills her either to solve the problem of the pregnancy or to punish her for her sexual excesses. Sometimes he announces his murderous intentions to her, and we hear her pleas for mercy. After stabbing, shooting, or beating her to death he disposes of her body in a shallow grave or a river. Finally, he admits his crime and pays for it with his life, either through a legal proceeding that culminates in execution or through a confrontation with the devil. Some of these ballads can be traced to English and Scottish models, while others are American in origin, though clearly patterned after British traditional and broadside ballads. (6) Tides such as "Knoxville Girl," "Omie Wise," "Pearl Bryan," "Little Sadie," "Florella," "Rose Connoley," "Banks of the Ohio," and "Pretty Polly" are perhaps familiar to many, since they remain part of the repertoire of contemporary folk and traditional musicians. Their stories are brutal and timeless; certainly, the specter of a man murdering his pregnant lover still looms large for contemporary Americans. (7)
How might one perform these ballads with their misogynist and violent messages without seeming to condone or endorse them? (8) Of course, the easiest solution is not to perform them at all. After all, there are hundreds of compelling ballads and folksongs to choose from. However, murder ballads have occupied a central role in American folk repertoire for over two hundred years, and refusing to perform them denies their significance and their power to speak to people then and now. Moreover, for many performers and listeners the tunes of these ballads are beautiful and ultimately not expendable. Thus, I would like to examine the ways one might respond to this challenge by focusing on performances of one of the most well known examples, "Pretty Polly."
The text of "Pretty Polly" can be traced to two English sources, Child ballad #4, "Lady Isabel and the Elfin Knight," and the British broadside ballad "The Gosport Tragedy," which is also known as "The Cruel Ship's Carpenter." (9) In both ballads a young woman is lured by a man to go away with him on the promise of marriage. When the elf/man in "Lady Isabel and the Elfin Knight" tells Isabel to take her clothes off so he can drown her in the river without ruining her fine garments, she asks that he turn his back out of respect for her modesty. When he does this she shoves him into the river, drowning him and saving herself. Things do not go so well in "The Gosport Tragedy" for the young pregnant Molly, or sometimes Polly, who is lured away, murdered, and buried by her lover. Sometime later, when he is aboard ship, her ghost, carrying the ghost of her baby, appears on the ship, and he is revealed as a murderer. The song "Pretty Polly," like many American versions of British ballads, is a condensation of the text's origins. The supernatural elements are typically stripped away, leaving only the broad outline of the formulaic murder plot and incorporating a couple of stark new lines and images: "a new-dug grave with a spade lying by" and "I dug on your grave the biggest part of last night." In American versions the woman's name is fairly consistently Polly, and her murderer's is Willie. The form of most American versions is also abbreviated--the typical four-line stanza of the British ballad is often reduced to a two-line stanza with one line repeated, resulting in an HB form. (10) The simple duple-meter tunes of "Pretty Polly" also differ from the compound duple-meter ones of the British models. Finally, American performances of this tune were often accompanied by the fiddle or banjo during the nineteenth and early twentieth centuries and later by the guitar. (11)
I would like to begin by briefly examining the traditional performance aesthetic of early recordings of "Pretty Polly," since they establish a kind of touchstone against which we can evaluate later performances; further, these recordings are what contemporary performers are, in effect, covering in contemporary renditions. (12) One of the most easily accessible recordings of "Pretty Polly" is by Estil C. Ball; this recording is representative of the typical text and tune of American versions of the ballad as well as the way it might have been sung and played by old-time musicians in the southern Appalachians. Bali's version, recorded in Rugby, Virginia, by John Lomax in 1941, became well known through Ruth Crawford Seeger's transcription of the song included in Lomax's collection Our Singing Country: Folk Songs and Ballads. Lomax later included this version, with piano accompaniment by Ruth and Charles Seeger based on Bali's guitar playing, in his collection Folk Song U.S.A.: 111 Best-Loved American Ballads. (13) Despite the influence that Bali's version undoubtedly had on later singers, there is some evidence that Ball based his version on a 1927 recording of B. F. Shelton from Kentucky (see example 1). (14) Although Ball used a guitar accompaniment, the constant sixteenth-note punctuations, insistent rhythms, and melodic riffs recall Shelton's banjo figuration, an accompanimental style more typically found in early performances of old-time music. (15)
[ILLUSTRATION 1 OMITTED]
Of even greater significance for the present study of performance style, however, is the singleness of affect that Ball and Shelton bring to the tune. Their dispassionate vocal delivery is conventional for traditional ballad singers, some of whom state explicitly that they intentionally perform in this manner. According to Almeda Riddle, a ballad singer from Arkansas,
You have to put yourself behind the song. By that I mean get out of the way of it. Present your story, don't perform it.... The difference between our most popular "folksingers" and me, they do perform and put too much of themselves into it. I lust get behind it. I don't want any of Almeda Riddle there. Let's get the picture of Mary Hamilton, the weeping, betrayed girl, before the public. And if your ballad is good enough, it'll hold them without anything that you do. You don't have to put any tricks to your voice or anything else, if you sing it with feeling. I do believe a ballad should be sung with feeling and with understanding. You must feelingly present a ballad or a hymn or spiritual. Now, if you want to perform on something like "The Three Nights Drunk" ... or something like that, if the song is an entertaining song, that's all right. Go right ahead. But these others, these classics, I do feel strongly about them. (16)
The significant distinction she makes here between ballads and entertaining songs has also been defined by scholars who differentiate ballads like "Pretty Polly" from lyric songs such as "Black Is the Color of My True Love's Hair," for example. G. Malcolm Laws reminds us that the action in ballads is "dramatized from beginning to end." Conversely, in lyric songs "the narrative and dramatic elements are weak. [The events are] reflected upon in the manner of soliloquy rather than dramatized. The story is not so much told as implied or suggested. Thus setting aside the matter of subjectivity, we may say that in contrast to songs, ballads dramatize their central events. Moreover, the primary purpose of the ballad is to tell a story, while that of the song is to express a state of mind." (17) Thus, when traditional singers perform a classic ballad, as reported by Riddle, they maintain an impersonal stance that does not portray a specific emotion. Similarly, ballad singer Sheila Kay Adams describes her role as a "dispassionate narrator" comparable to that of a "disinterested newspaper writer reporting just the facts" of the story. (18) Greil Marcus writes eloquently of this remote affect Dock Boggs brought to his 1927 recording of "Pretty Polly":
There is a supernatural tinge to the song as it emerges from Boggs's performance, even though nothing unearthly is named or even hinted at; the sulfurous odor comes up because as Boggs tells it, there seems to be no will in the story, only fate, or ritual.... Boggs drags out his vowels, this time every one on a flat plane. The man who is telling this story ... knows what is waiting at the end of every line. There are no surprises, there is no possibility of surprise. That's what is so murderous about the performance. (19)
Commenting specifically on Shelton's recording of "Pretty Polly," Bertrand Bronson states,
The dominant impression conveyed by a good folk-song sung in the best traditional style is, it cannot be too strongly insisted, one of genuinely classic impersonality.... [They] are starkly powerful in their reserve and understatement.... One of the finest illustrations known to me is the rendition on an old Victor phonograph record ... by a folk-singer named B. F. Shelton.... The resulting effect of distancing is comparable to the removal, in a classical drama, of immediate catastrophic action from the stage, and its consequent reduction to a messenger's recital.... [T]he characteristic singing ... is of an incantatory masklike aloofness that apparently makes no concessions to ordinary fluctuations of human sympathy or excitement, nor shows any awareness of audience, hut tacitly acknowledges, throughout, its allegiance to a higher court, the strangely abstract, impersonal law of tradition. (20)
Many contemporary singers who have immersed themselves in old-time music emulate this traditional style of performance, as Riley Baugus does in his 2001 recording. (21) One is immediately struck by his rather impersonal manner, marked by a quick tempo, lack of dynamic changes or tempo shifts, virtuosic banjo figuration and its unadorned repetition of the melody during instrumental breaks, and, in particular, the detached affect. The result, however, is that this emotional distance and hard-driving banjo accompaniment might be heard by some contemporary listeners as a heartless acceptance of the ballad's violent scenario. Indeed, Laws states, "Because of their objective style the Child ballads may seem unsympathetic or cold blooded in their handling of tragedy," although he also reminds us that the ballads "actually, of course, ... seethe with feeling, but they mask emotions under an exterior of reserve." (22) Certainly, traditional ballad singers speak of their own emotional responses to hearing these ballads all their lives, and not all contemporary listeners will experience an aloof performance as one that accepts the violence. (23) Some may hear this dispassionate style as only representing another era's set of traditional aesthetic values, while others may hear it as a kind of numbed response to the violent scenario. Nevertheless, for some listeners this performance style does allow the possibility that the violent text seems to remain unchallenged, particularly if the performance or recording is not accompanied by remarks or liner notes that contextualize this aesthetic or ask the listener also to maintain a distant stance with regard to the violence. This detached style, replicated in dozens, if not hundreds, of contemporary ballad performances, can be unnerving in a manner reminiscent of that described by Mark Franko and Annette Richards in their study of historical performances: "If performance is understood as 'restored behavior,' as fundamentally repetitive or reiterative, ... it necessarily brings back the past to unsettle the present." (24)
Ultimately, the performer who brings back a murder ballad from the past and who wishes to sing it without seeming to endorse or reinscribe its violence may have to abandon this distanced position to some degree. Through disrupting this seemingly "natural" or "timeless" aesthetic the performer can take a resisting position that offers a new way of singing, hearing, and understanding these brutal stories. (25) Of course, there is no one way to do this resisting, but each resisting performance becomes, in the words of Adrienne Rich, a "re-vision: the act of looking back, of seeing with fresh eyes, of entering an old text from a new critical direction." (26) How might one begin to devise a resisting performance of a murder ballad such as "Pretty Polly"? I would suggest that in the case of the murder ballads that are told from the first-person perspective of the murderer a resisting performance would be one that gives a voice to the woman's experience without merely highlighting or enacting the murder, which is usually far from muted in these songs. But this is easier said than done. Judith Fetterley remarks in her study of women as resisting readers, "Women obviously cannot rewrite literary works so that they become ours by virtue of reflecting our reality." (27) While it might be an enjoyable exercise to rewrite the texts of these ballads and songs, this choice sidesteps the issue of performing the traditional ballads themselves. Indeed, much more so than the tunes it is the texts of these songs that give them their particular identity, and it is the texts that have garnered the lion's share of attention by both scholars and performers alike. Paradoxically, however, scholars and traditional singers usually assert that a ballad isn't a ballad until it is sung and that it is the music of ballads that gives them their power. (28) Susan Cook, in her study of the ballad "Fuller and Warren," reminds us:
The ballad's tune is a crucial element in the narrative transaction, acting as a powerful fixative, sticking, as it were, the story in the minds of the singer and listener and aiding in later recall. Furthermore, the musical nature of the ballad emphasized its particular communal function as entertainment. Ballads were regularly repeated in a community as a diversionary activity, and this repetition provided individual stories with more opportunities to teach their codes of behavior. Indeed, in the case of "Fuller and Warren" the ballad's ideological interpretation, when set to music, outlived other judgments of the events. (29)
I would therefore suggest that the music itself, not the text, is the strongest potential site of resistance and our best opportunity to hear Polly's voice. Accordingly, the performance choices made primarily with regard to the music, not the text, are the subject of this essay.
The great number of commercial recordings of "Pretty Polly" provide a wide range of performance choices that allow an investigation into the problem of resisting rather than endorsing the violence against women. These performances run the gamut from those that are traditional in style, instrumentation, and tempo to those that rely very little on traditional performance practices of the ballad and instead depend on musical choices designed to appeal to a contemporary audience that may not be moved by banjos and hard-driving, impersonal vocals. I want to consider several performances along this continuum before arriving at and exploring what I believe is the most successful resisting performance I have encountered.
The first example is by Lily May Ledford, a singer and banjo and fiddle player from Kentucky. (30) During the 1930s she was the star member of the first all-girl string band, the Coon Creek Girls. Ledford's performance is traditional in every way: she performs the ballad at a quick pace with a straightforward singing style and a hard-driving, old-time banjo accompaniment that does not waver from the song's tune. Nothing about her performance of the song could be considered resisting. But what I find interesting about Ledford's rendition is the anecdote she uses to introduce the song in live concerts:
"Pretty Polly" was one that Mama told us we could sing. Now "Wild Bill Jones," she put her foot down on that, and "Little Cory" and "Frankie and Johnnie" 'cause of drinking, and wine drinking, you know, and moonshining mentioned in some of 'em. And she said we could sing "Pretty Polly," and she mentioned some other old songs. And I said, "Well Mama, I don't know why you think those songs that's got whiskey and drinking in 'em, that don't mean we have to go out and bring it in here in the house and drink it, that just mean we be singing about it. Everybody else does." She said, "Yeah, you want to be like that sorry bunch that lives down the road, all the women down there a-pickin' the banjar and singing them old drunkard songs and," she said, "the men a-making rot-gut moonshine." And she said, "Now that's alright maybe for the boys, but," she said, "now I don't want you girls to sing that. Now "Pretty Polly' is one, there's you a good love song." I said, "Mama, 'Pretty Polly' has got a murder in it. That man killed Pretty Polly. Is it better to kill somebody or to drink?" And she said, "I think it's better to kill somebody"; she said, "He probably wouldn't a killed her if he hadn't been drunk." (31)
While Ledford uses this humorous story to a charming effect, through it she also shrewdly calls attention to Polly's murder and places it firmly within a context of poverty, drinking, and violence. This maneuver, I believe, is a small step toward resisting the inevitability of Polly's murder.
A recording of "Pretty Polly" that spent three weeks at the top of the bluegrass charts in 1997 is by bluegrass legend Ralph Stanley with Patty Loveless. (32) This version abandons the traditional old-time accompaniment of solo banjo for a full bluegrass band. (33) Nevertheless, the affect in almost every other way remains traditional, with its fast tempo and straightforward vocal style. But what makes this rendition of interest is that Stanley and Loveless perform the song as a duet, taking on the roles of Willie and Polly. This choice seems natural; "Pretty Polly" includes dialogue between the two characters, and what better way might there be to represent their interchange than to hear the male and female voices of the performers standing in for the murderer and the murdered. However, this dialogue does not guarantee a resisting performance if by that we mean one that gives voice to Polly's experience. There is an impediment that prevents the duet structure from giving Polly a voice. Polly's words are not her own but are Willie's. The song is clearly a first-person ballad, with Willie as the narrator who sometimes speaks as Polly. He refers to himself outside of dialogue passages; she does not. And after Polly's death the ballad continues. Willie is still alive to comment; Polly is not. (34) Thus, Polly's words are not of her own experience but are Willie's words about her experience. Musicologist Ruth Solie calls this type of maneuver a "spurious autobiographical act," one that is "the impersonation of a woman by the voices of male culture." (35) Merely putting the words that Willie has Polly speak into the body of a woman singer does not give Polly a voice. In fact, it extends the impersonation, simply dramatizing the narrative that Willie constructs. This is why, when Patty Loveless sings as Polly, she does not necessarily need to deviate from the impersonal vocal style of a traditional ballad singer. (36) Like a solo ballad singer, Loveless sings Polly's lines from a distant viewpoint, maintaining the ballad's traditionally detached representation of the murder.
In an attempt to make these ballads speak to contemporary audiences, performers often abandon a traditional performance aesthetic, as Judy Collins does in her 1968 recording. (37) In her transformation of the song from an old-time or bluegrass tune into a kind of folk-rock power ballad, Collins saturates her rendition with pathos, presumably as a comment on the unjust and horrible murder of Polly. First, she chooses a very slow tempo, half the rate of a typical traditional performance, causing her occasionally to interrupt the narrative with a breath in the middle of a line. Making up for the relentless forward motion lost with this slow tempo, Collins builds tension through the song by increasing the intensity gradually with each new verse. Amplified instruments and drums are added one at a time to the solo acoustic guitar of the first verse. The dynamic level grows with this fuller texture, and the increasingly insistent backbeat pushes the narrative inexorably to its conclusion. Simultaneously fueling this drive to the culmination of the story, Collins's vocal line and the instruments' countermelodies deviate more and more from the tune as the song progresses, varying it until the final line about Willie "running away," which receives her most prolonged embellishment. These musical choices allow her to linger on the details of the story, wringing the maximum expression out of each verse. What is particularly interesting is that she dramatizes the narrative's dialogue by assuming a different vocal style for the two characters in a sort of duet for one. She sings many of Willie's words in a very heavy chest voice with no vibrato, using this timbre to great effect on the words grave and flow in the lines "digging on your grave for the best part of last night" and "he stabbed her through the heart and her heart blood did flow." Conversely, when Polly pleads for her life Collins adopts a very breathy sound that calls for a sudden drop in volume from the band. She also uses this lighter timbre for the stanza about the wild birds weeping over Polly's body as Willie buries her. Collins returns, however, to Willie's chest voice at the lines on paying a debt to the devil.
This is a version that a mainstream audience might well enjoy; its popular style, instrumentation, and recognizable genre are familiar to many listeners who would probably be compelled through this version to hang on every word of the story and sympathize with Polly completely. But the question remains: does this performance resist the murder or merely engender pity for the murdered girl? Despite her heartrending performance I believe the ultimate effect of Collins's performance is a dramatization of the murder rather than a performance that takes a stance against the violence. In fact, the very elements that Collins uses to such great effect to engender pity--the different vocal timbres for the two characters and the overwrought emotional style--are what prevent Collins's performance from resisting the violence of the song. Rather, this kind of performance seems more to savor the murderous story, even to wallow in it. Collins's performance encourages listeners to feel Polly's pain, and, I would argue, they are urged even more to feel Collins's pain in feeling Polly's pain. Collins seems to be trying awfully hard to empathize with Polly, and in so doing she inserts herself smack in the middle of the ballad, exactly where traditional singer Almeda Riddle said she would be. Recall Almeda's words: "The difference between our most popular 'folksingers' and me, they do perform and put too much of themselves into it." (38) There is one other effect that I worry this performance may have. Because Collins in no way adopts a traditional performance style in vocal production, instrumentation, or genre, she not only reaches out to a mainstream audience (in itself not a bad goal) but she also, intentionally or not, distances herself from the original time and context of the ballad. In so doing she runs the risk of seeming to take a superior stance vis-a-vis the culture from which this song comes. Thus, her listener may feel secure in the belief that it was in an earlier and more brutal time and place that Polly was killed in an era and region that no longer exist or matter, except when we wish to update them in a way that we can enjoy. If this is the final outcome of the performance, I would maintain that it cannot be called a resisting performance, as I explain more fully at the end of this essay.
The previous two examples attempt to give Polly a voice by locating it in the dialogue passages of the song, either through a duet or through a dual use of vocal timbre. The next rendition of "Pretty Polly" abandons this strategy, putting Polly's voice into a wordless countermelody sung by someone other than the principal singer of the text. Hilarie Burhans's 2003 version straddles a traditional and contemporary performance aesthetic. (39) Like Lily May Ledford, she begins and sings much of the song with only a banjo accompaniment. However, she establishes a slow tempo through which to tell the story, and her breathy vocal timbre from the outset and throughout the song seems to represent Willie's menacing and seductive voice. Conversely, Polly's voice is represented by wordless melismas that begin when Polly speaks for the first time: "Oh, Willie, oh, Willie, I'm afraid, afraid of your ways / I'm afraid you mean to lead my poor body astray." By synchronizing the countermelody's entrance with Polly's first utterance, Burhans clearly establishes that Polly's voice is located outside the text and even the tune. Burhans's ostensible resisting strategy is to end with Polly's fear rather than her death, since she concludes her performance with the verse in which Willie affirms Polly's suspicions by saying, "Polly, my pretty Polly, girl you guessed it just about right / I been digging on your grave here for the best part of last night." As she explains in the liner notes for this song, Burhans "left out some of the more gruesome verses ..." (40) This tentative and evanescent nod toward the murder and her omission of the murderous verses reveal the primary way she responds to this song. She averts her and our eyes from the murder. Like a skittish rubbernecker who quickly looks away at the last second from the scene of a gruesome accident, this performer does not really confront the grisly scene. In a sense, by denying the reality of the murder while highlighting its existence, Burhans takes an ambiguous stance that ultimately is not resisting at all. Rather, this version allows us to enjoy the song up to a point, hearing Polly's fear through the countermelody but abandoning her to her fate in our refusal to listen to the final outcome. As much as I believe Burhans intended her performance to resist the violence, as demonstrated by her liner note remark, I also believe she fails to resist what she cannot sing.
Although I find Burhans's performance unsuccessful as a resisting performance, I do believe that she was on the right track when she located Polly's voice outside the text and the tune of the traditional song, as my final example by the New Coon Creek Girls does. This all-woman bluegrass band, named in honor of Lily May Ledford's group from the 1930s, recorded "Pretty Polly" in 1991, and in many ways their version is quite traditional. (41) The singer maintains an unaffected vocal style, and the entire text is sung, including the murder and the retribution. Moreover, they abandon their usual bluegrass style, and only two members of the band perform, one on vocals and banjo, the other on fiddle, recalling the standard old-time instrument arrangement. In the context of their album of contemporary bluegrass numbers this old-timey performance of "Pretty Polly" can be heard as "authentic." As a result, their traditional performance style does not distance them or their listeners from the story but instead manages, somewhat paradoxically, to bring the characters to life, rather like an old sepia-toned photo might. We expect Polly's world to sound ancient, and it does in this version; indeed, on first hearing it may even pass for a traditionally detached, nonresisting performance. But this seemingly unmediated quality is, I believe, what helps them construct a resisting performance. The standard old-time duo of banjo and fiddle introduces a powerful medium, the fiddle, through which Polly's voice can be heard.
I want to make two observations concerning the choice of the fiddle to represent Polly in this ballad. In the context of old-time music the fiddle has many associations. The long-standing link between the fiddle and the devil is used to great effect in this performance. (42) The fiddle itself seems to dole out the retribution in the final stanza as Willie pays his debt to the devil. But even more intriguing is the association that the fiddle has with the bodies of women, particularly those who have died. Of course, there is the film The Red Violin, in which the blood of a young woman who tragically dies in childbirth is lovingly used by her husband to create the varnish for a violin he makes. The film traces the violin as it sings the melody and spirit of the woman through three centuries. There is also a nineteenth-century tradition of fiction in which the violin assumes the voice of a woman who has died or, in the case of murder, reveals the name of her killer. (43) Of even greater resonance for our purposes is the example from the traditional murder ballad "The Twa Sisters." In it a woman drowns her sister when she discovers that a young man has been courting them both. Eventually, the sister's body is discovered by a miller, who uses her bones and hair to fashion a fiddle that reveals the name of her murderer. (44) While "Pretty Polly" does not end with Polly's ghost revealing her murderer, the final moments of the New Coon Creek Girls' rendition suggest such a reading when the fiddle slips out of the mode at the end and wavers on otherworldly sounding harmonics. The use of the fiddle to represent not only Polly but also the devil and perhaps even Polly after her death reintroduces a supernatural element to the ballad that is rarely if ever heard in its American versions. In view of these cultural references it seems particularly astute that the New Coon Creek Girls would use the fiddle to speak for Polly.
In this performance the fiddle plays fragmented responses and countermelodies at the end of line 1 and during line 2 of each stanza. It also plays more expansive melodic interludes between stanzas (see table 1). The middle column of table 1 indicates the lengths of the phrases, responses, and interludes. For example, in the first stanza each line is eight beats long; the first response (indicated by italics) is six beats; the interlude moving to stanza 2 is ten beats. The third column summarizes the character of the melodic fragments and interludes of the fiddle. By establishing the fiddle as a melodic response to the tune and text of "Pretty Polly," the New Coon Creek Girls cast the fiddle in a role that can be understood to be commenting for Polly, not about her, as Willie does. Thus, the fiddle, which responds in counterpoint to the text, is free to reflect, from Polly's point of view, on all the words and actions within the ballad, both Polly's and Willie's.
As the ballad begins we hear the somewhat embellished but very recognizable melody of "Pretty Polly" played in its entirety on solo banjo. In the first stanza the banjo accompanies the voice in an unadorned fashion, with just a bit of filler between the lines and only the slightest hint of a melodic interlude after the stanza. Significantly, the fiddle is initially heard in the second stanza in direct response to Willie's question: "Oh, where's Pretty Polly? Over yonder she stands." Clearly representing Polly, the first melodic response has a somewhat docile character, a sort of musical curtsy. Mirroring the contour of the melody, this response is in the lower register of the mode, while the second fragment moves to a higher range and is enlivened a bit with syncopation. At the end of this second stanza the fiddle plays an interlude that becomes associated with Polly; this melody moves within the low range of the mode, highlights a scalewise passage from the dominant to the tonic, and is rhythmically unsyncopated (see example 2). In short, the fiddle's opening gestures establish a sort of "Polly identity" that will be recalled and altered as the song progresses.
[ILLUSTRATION 2 OMITTED]
The fiddle maintains this "Polly identity" in the third stanza. Although the melodic fragments are different, particularly with the introduction of double-stops, the gestures remain consistent, moving from low to high. The closing interlude is a reprise of the one already heard. However, things begin to change in the fourth stanza, in which Polly begins to sense that something is not right: "I led her over mountains and valleys so deep / But then Pretty Polly, she began to weep." The two melodic fragments at this point remain in a low register, falling entirely below the range of the earlier ones. A shift is also heard in the interlude (see example 3). It begins as the previous ones did but quickly loses momentum, coming to a curtailed end and sinking an octave below the range of the song. Significantly, the last pitch of this interlude seamlessly elides with the pitch of the singer's entrance in the fifth stanza, which introduces the first words Polly utters in the text: "Oh, Willie, I'm afraid of your ways / I'm afraid you are going to lead me astray." This maneuver further links the fiddle with Polly. In the fifth stanza the melodic fragments in the fiddle begin to reflect Polly's growing fear. The first recalls the opening motif that introduced her; however, now the motif is a fifth higher and enters on a pitch that is dissonant with the chord at that point. The second fragment remains in the upper region, and the dissonance continues through a series of sighlike figures or appoggiaturas (see example 4).
[ILLUSTRATIONS 3-4 OMITTED]
While the fragments of the fifth stanza began and remained in the upper register of the mode, indicating Polly's growing agitation, the sixth stanza inverts the established low-to-high progression for the fragments as Willie confirms Polly's fears. Even more significant is the interlude that follows Willie's announcement: "I dug on your grave the biggest part of last night" (see example 5). The fiddle runs through scale-wise passages, slides between pitches, covers the span of two octaves within a few moments' time, and punctuates the melody with numerous dotted and syncopated rhythms. Polly's melody is subtly embedded in this extended interlude, but it is surrounded by figuration that effectively reveals Polly's panic as she is about to be murdered.
[ILLUSTRATION 5 OMITTED]
As Polly pleads for her life in stanza 7 the fiddle recalls the transposed fragments and appoggiaturas of stanza 5 but this time adds a double-stop at the interval of a seventh, heightening the tension. And, like stanza 5, both melodic fragments remain in the upper range of the mode. In the interlude that follows her actual murder--"But into her bosom I plunged the fatal knife"--the fiddle repeats the seventh but without any softening appoggiaturas. Rather, the dissonance stabs four times in syncopation and concludes on an open fifth (see example 6).
[ILLUSTRATION 6 OMITTED]
As Polly dies in the eighth stanza the fragments become simpler, covering the narrower range of a fifth, and, although they revert to the normal low-to-high progression, the double-stops disappear. The fiddle's most vivid moment occurs in the short interlude following Willie's words, "I left Pretty Polly for the birds to weep and moan"; there it clearly suggests the birds' cries with high short trills (see example 7). (45)
[ILLUSTRATION 7 OMITTED]
The final stanza and conclusion of this performance have a decidedly supernatural tone that is relayed entirely by the fiddle. As Willie sings, "And now to the devil a debt I must pay," the fiddle no longer simply plays melodic fragments; rather, it sets up a syncopated rift that continues insistently throughout the entire stanza, first on open fourths and then dissonant seconds in the second phrase. Polly's fiddle is now transformed and immediately conjures up familiar images of the devil himself as a fiddler (see example 8). The interlude that follows recalls Polly's melody one last time before the banjo and fiddle slip into a final repetition of the now textless tune of the ballad. The fiddle dominates the sonority here, sliding between pitches, embellishing notes, and inserting double-stops as it did throughout the song, even tacking on one last iteration of Polly's melody at the end of the tune. And the fiddle has the last poignant word when the performance is presumably over. Upon reaching the tonic the fiddle slides down a whole step and then hovers above the scene in a ghostly shimmer of harmonics completely outside the range and mode of the piece.
[ILLUSTRATION 8 OMITTED]
Working in tandem with the fiddle responses is the improvised quality of the performance. In all but two stanzas the standard phrase lengths established in stanza i are altered, usually by elongating a word or more often a name in the second line, as indicated in table 1 by italics. (46) These phrase extensions begin with the second stanza, just after the fiddle is introduced. In stanzas 2 and 3 the second phrase is elongated by four beats; however, at the first sign of trouble, stanza 4, the standard phrase lengths reappear, propelling the narrative forward a bit faster. Stanzas 6 and 7 are of particular interest. At the end of stanza 6 Polly's panic is expressed in the longest interlude of thirty-six beats. Conversely, the shortest interlude occurs at the end of stanza 7 at the moment Polly is stabbed. At only six beats long, the attack on Polly is swift, and the performers do not linger over it. Neither do they dwell on her death in stanza 8, whose phrase lengths are not elongated.
What makes the New Coon Creek Girls' version effective as a resisting performance is that by manipulating the pace of the action and by closely aligning the fiddle melodies with Polly's experiences the performers draw the listener's attention to Polly's feelings of suspicion, dread, and terror, but her actual murder is not accentuated. As a result, the listener is denied the opportunity to savor the violence itself while simultaneously prevented from disregarding it. Further, the elements of instrument choice, melodic character, and musical form all work together to create a dynamic emotional counter voice to that of the narrator in a move that recalls Judith Fetterley's approach: "Feminist criticism represents the discovery/recovery of a voice, a unique and uniquely powerful voice capable of canceling out those other voices, which ... spoke about us and to us and at us but never for us." (47) In their performance the New Coon Creek Girls discover a voice for Polly in the fiddle, a voice separate from that given to her through the words Willie sings. Thus, recalling the distinction made earlier between the impersonal ballad and the expressive lyric song with its depiction of a character's emotional state, I would argue that when the New Coon Creek Girls give Polly a voice that is representative of her emotional, not verbal, response to Willie's threats they arrive at a resisting reading. Accordingly, one way to create a resisting performance of a narrative ballad is to introduce an emotional perspective usually found in lyric song without altering the words significantly, dramatizing the dialogue, or relying on exaggerated vocal displays.
I would further propose that the most powerful and effective resisting performances of a ballad like "Pretty Polly" are those in which many of the song's traditional or original elements are retained, in a sense taking on the violence on its own terms and in its own context. In working out strategies for resisting readings of texts from other eras and contexts, Fetterley posits, "While women obviously cannot rewrite literary works so that they become ours by virtue of reflecting our reality, we can accurately name the reality they do reflect and so change literary criticism from a closed conversation to an active dialogue." (48) I believe her approach is applicable to resisting performances as well. Resistance to a traditional song, or any song from outside of our present context, does not necessarily come about by performing it in a way that reflects a contemporary reality or aesthetic. Indeed, performing a traditional ballad in this way runs the risk of being misunderstood as a parodic or patronizing rendition. Rather, the possibility for real resistance and dialogue emerges when a performer explicitly works within the reality that the song reflects and within the context from which the song comes. (49) In the case of "Pretty Polly" the reality the song reflects most clearly is nineteenth-century Appalachia, a context that some contemporary listeners associate with the old-timey music of banjos and fiddles as well as with violence, ignorance, misogyny, and primitiveness. Consequently, for some listeners, such as those who inhabit my first scenario, to sound Appalachian is to sound violent, ignorant, misogynist, and primitive. This link between musical sound and cultural character can be so strong that it may seem that the only way to break out of or resist the violence is to abandon or resist the sound. Thus, when a singer like Collins, for example, performs "Pretty Polly" outside an Appalachian sound context she seems to be saying that we can only resist this violence when we don't sound Appalachian (when we aren't Appalachian?). But how much more powerful it is to resist the violence and remain Appalachian. A traditional and resisting performance of "Pretty Polly" does not simply write off the violence as coterminous with its Appalachian context. Rather, it suggests the possibility that Appalachia is not automatically a place where violence and misogyny are accepted. Indeed, such a resisting performance belies that stereotype and allows the listener in my second scenario to reclaim a beloved musical sound. As Edith Randam Rogers explains, "Whether a popular ballad lives or dies depends not so much on the quality of its ancestors as on its ability to evoke new associations that touch the values and concerns of successive generations." (50) The New Coon Creek Girls provide a model of this approach with their resisting performance that expresses Polly's emotional experiences in the musical language and dialect of the song's American origins, manifested through a quintessential old-time instrument, the fiddle.
I would like to thank Heather Buchman for her many helpful suggestions and for transcribing the musical examples of "Pretty Polly" as performed by the New Coon Creek Girls. Thanks also to Nancy Sorkin Rabinowitz and Peter J. Rabinowitz for their valuable comments on early versions of the essay. I would also like to acknowledge my student Annie Neikirk for preparing the musical examples. An earlier version of this paper was presented at the International Association for the Study of Popular Music held at the University of Virginia, Charlottesville, October 2004.
(1.) I am aware that I am speaking in broad generalities here. All old-time music is not Appalachian, and bluegrass music is a commercial genre that, while created in Appalachia, is now played and heard in all regions of the United States and throughout the world. Nevertheless, in the popular imagination these genres are consistently conflated together and with the Appalachian region.
(2.) Ronald D. Eller, foreword, in Dwight B. Billings, Gurney Norman, and Katherine Ledford, eds., Confronting Appalachian Stereotypes: Back Talk from an American Region (Lexington: University Press of Kentucky, 1999), ix. Although this valuable collection of essays does not include a discussion of Appalachian music, in his introduction to the volume Billings quotes a trenchant example of the Appalachian musical stereotype. About the Swedish band Rednex an Entertainment Weekly music critic wrote: "Nearly every song is sung in the voice of a drooling, bug-eyed inbred from some imaginary Appalachian trailer park.... Destined for a Scandinavian version of Hee Haw--where they belong" (4). For other resources on Appalachian cultural stereotypes see Robert J. Higgs, Ambrose Manning, and Jim Wayne Miller, eds., Appalachia Inside Out, vol. 1, Conflict and Change, and vol. 2, Culture and Custom (Knoxville: University of Tennessee Press, 1995); and W. K. McNeil, ed., Appalachian Images in Folk and Popular Culture, 2nd ed. (Knoxville: University of Tennessee Press, 1995).
(3.) I would suggest that this position led to the popularity of the film O Brother, Where Art Thou? even though the film does not take place in Appalachia and actually features very little bluegrass. The soundtrack went quintuple platinum, although it received very little airplay on country music radio. See Gary Dretzka, "Satellite Radio to the Rescue," http://www.salon.com/tech/feature/2002/06/019/ satellite_radio/, accessed 18 January 2005.
(4.) "Ballad tunes may work in much the same way that Catherine Clement says operatic scores do, as transitional objects that allow the stories to pass unnoticed" (Susan C. Cook, "'Cursed Was She': Gender and Power in American Balladry," in Susan C. Cook and Judy S. Tsou, eds., Cecilia Reclaimed: Feminist Perspectives on Gender and Music [Urbana: University of Illinois Press, 1994], 202-24, 216).
(5.) The following studies provide a start for an examination of the murder ballad genre: Anne Cohen, Poor Pearl, Poor Girl! The Murdered-Girl Stereotype in Ballad and Newspaper (Austin: American Folklore Society/University of Texas Press, 1973); Olive Woolley Burt, American Murder Ballads and Their Stories (New York: Citadel Press, 1964); D. K. Wilgus, "A Tension of Essences in Murdered-Sweetheart Ballads," in James Porter, ed., The Ballad Image: Essays Presented to Bertrand Harris Bronson (Los Angeles: Center for the Study of Comparative Folklore and Mythology, University of California, Los Angeles, 1983), 241-56; David Atkinson, "Magical Corpses: Ballads, Intertextuality, and the Discovery of Murder," Journal of Folklore Research 36, no. 1 (1999): 1-29; and David Atkinson, "Magical Corpses and the Discovery of Murder," in his The English Traditional Ballad: Theory, Method, and Practice (Hams UK: Ashgate Publishing, 2002), 185-232.
(6.) I use the term ballad in reference to those songs that tell a story; this category includes the Child ballads (those collected by Francis James Child), British broadside ballads, and American ballads either newly composed or based on English models. These distinctions are outlined in G. Malcolm Laws, Jr., American Balladry from British Broadsides: A Guide for Students and Collectors of Traditional Song (Philadelphia: American Folklore Society, 1957), 1-2; and G. Malcolm Laws, Jr., Native American Balladry: A Descriptive Study and a Bibliographical Syllabus, rev. ed. (Philadelphia: American Folklore Society, 1964), 1-2.
(7.) As I write this essay we hear regular reports of the Scott Peterson trial and of the murder of Lori Hacking, to which her husband, Mark, has confessed; both women were pregnant, both presumably were killed by their husbands.
(8.) While this essay focuses on the resisting performer, one can also respond to these murder ballads as a resisting listener, as Cook urges in her study of another murder ballad: "We must become resisting listeners, adding our textual ripostes and dissonant countermelodies to these same old songs" ("Cursed Was She," 217). Suzanne G. Cusick explores resisting performances in the classical music realm in her "Gender and the Cultural Work of a Classical Music Performance," repercussions 3 (1994): 77-110.
(9.) Versions of "Lady Isabel and the Elfin Knight" can be found in Helen Child Sargent and George Lyman Kittredge, eds., English and Scottish Popular Ballads Edited from the Collection of Francis James Child (Boston: Houghton Mifflin, 1904), 4-7; Bertrand H. Bronson, The Traditional Tunes of the Child Ballads, vol. 1 (Princeton N j: Princeton University Press, 1959), 39-100; Olive Dame Campbell and Cecil J. Sharp, English Folk Songs from the Southern Appalachians (New York: G. P. Putnam's Sons, 1917), 3-8; and Henry M. Belden and Arthur Palmer Hudson, eds., Folk Ballads from North Carolina, vol. 2 of The Frank Brown Collection of North Carolina Folklore, Newman Ivey White, gen. ed. (Durham NC: Duke University Press, 1952), 15. Versions of the text, tune, and history of "The Gosport Tragedy" can be found in David C. Fowler, "The Gosport Tragedy: Story of a Ballad," Southern Folklore Quarterly 43 (1979): 157-96; G. L. Kittredge, ed., "Ballads and Rhymes from Kentucky," Journal of American Folklore 20, no. 79 (1907): 251-77; Belden and Hudson, Folk Ballads from North Carolina, 234-40; Dorothy Scarborough, A Song Catcher in Southern Mountains: American Folk Songs of British Ancestry (New York: Columbia University Press, 1937), 126-34. For an extensive bibliographical guide to "Lady Isabel and the Elfin Knight" see Tristram P. Coffin, The British Traditional Ballad in North America (Philadelphia: American Folklore Society, 1963), 25-28. For a comparable guide to "The Gosport Tragedy" see Laws, American Balladry from British Broadsides, 268-70.
(10.) D. K. Wilgus refers to the American version of "Pretty Polly" as a blues ballad in his liner notes to The Doc Watson Family, Folkways Records Album No. FA 2366 (New York, 1964). In his longer study of the genre Wilgus writes, "The blues ballad does not so much narrate the events of a story as it celebrates them. The amount of explicit narrative varies.... [The blues ballad] text depends on a narrative referent, that is, an actual event occurring in a community or a traditional pattern such as 'the murdered sweetheart' of "the criminal brought to justice'" (D. K. Wilgus and Eleanor R. Long, "The Blues Ballads and the Genesis of Style in Traditional Narrative Song," in Carol L. Edwards and Kathleen E. B. Manley, eds., Narrative Folksong: New Directions; Essays in Appreciation of W. Edson Richmond [Boulder co: Westview Press, 1985], 435-82, see 442-43). He explains a possible misconception as well: "Despite the connotations of the term, the American blues ballad predates the blues form in American culture.... Blues ballad variants do occur in AAB text structures, but they ,seem late in the traditions of the ballads concerned.... The patterns of repetition in the American blues ballad occur in Black and White spirituals and in Black work songs. While we might look to them for origins, we can also view the formal structures as parallel developments. Certainly we can postulate an interaction between Black work and social songs and the White lyrics and fiddle songs" (Wilgus and Long, "The Blues Ballads," 458-59). Norm Cohen adds this to Wilgus's definition: "There is indeed a story implicit in the lines of such a composition, but, in blues fashion, the words comment upon the drama more than expose its details. The blues ballad is the editorial to the broadside ballad's front-page story. Sometimes the underlying story is not so deeply hidden" (liner notes to Going down the Valley: Vocal and Instrumental Styles in Folk Music from the South, New World NW 236 [New York, n.d.]).
(11.) Josiah H. Combs comments on the tune variations: "One of the best examples of the harrowing of the folksong in the hands of the banjo picker is 'The Gosport Tragedy,' commonly known as 'Pretty Polly.' The traditional airs of this song are strangely beautiful, but are hardly to be recognized when heard on the banjo" (Combs, Folk-Songs of the Southern United States, ed. D. K. Wilgus [Austin: American Folklore Society/University of Texas Press, 1967], 92). Charles K. Wolfe writes, "Instrumental traditions in nineteenth-century Kentucky were pretty much dominated by the fiddle and the banjo. It was not unusual to find a ballad singer accompanying himself on the fiddle, usually playing a melody line in unison with the singing. By the turn of the century, according to Combs, some banjoists were starting to adapt old ballads (such as 'Pretty Polly') to their banjo styles" (Kentucky Country: Folk and Country Music of Kentucky [Lexington: University Press of Kentucky, 1982], 11).
(12.) I should make it clear that I am using the slippery term traditional here to represent any performance in which the song is sung by a solo singer accompanied only by instruments such as the banjo, fiddle, and guitar, whether historical or contemporary.
(13.) John A. and Alan Lomax, Our Singing Country: Folk Songs and Ballads (1941; repr., Mineola NY: Dover, 2000), 172-73; John A. and Alan Lomax, Folk Song U.S.A.: 111 Best-Loved American Ballads (New York: Signet, 1947), 386-87. The 1941 recording of Ball singing "Pretty Polly" can be found on A Treasury of Library of Congress Field Recordings: Selected and Annotated by Stephen Wade, Rounder CD 1500 (1997). Ball recorded later but essentially identical versions in 1959: Voices from the South, vol. 1 of Southern Journey, Rounder CD 1704 (1997), and Bad Man Ballads, vol. 5 of Southern Journey, Rounder CD 1705 (1997); and in 1965: High Atmosphere: Ballads and Banjo Tunes from Virginia and North Carolina, Rounder CD 0028 (1995). All four of these recordings are available through the iTunes Music Store website.
(14.) In his liner notes to Bali's 1941 recording Wade writes, "Lomax's 1940 field notes mention Kentuckian B. E Shelton's 1927 Victor banio recording of 'Pretty Polly' as the source of Bali's arrangement of this well-known eighteenth-century English broadside" (9). Shelton's recording can be heard on RCA: THE BRISTOL SESSIONS, 1927, vol. 1, RCA Nashville 65 131 (2002). There seems to be additional evidence for a Kentucky origin of this version of the tune. Dock Boggs, from the western Virginia/eastern Kentucky region, also recorded "Pretty Polly" in 1927; this recording can be heard on Dock Boggs Country Blues: Complete Early Recordings (1927-29), John Fahey's Revenant (1997). His 1968 recording is on Dock Boggs: His Folkways Years, Smithsonian Folkways, SFW 40108 (1998). At least two editors included Kentucky sources of the tune in their collections prior to Lomax: John Harrington Cox, Traditional Ballads and Folk-Songs Mainly from West Virginia (repr. of his 1939 editions of pamphlets), 73-75--this version was collected in Kentucky in 1925, and although it is clearly the American version of "Pretty Polly," it retains a mention of the ship and Polly's baby found in "The Gosport Tragedy"; see also Combs, Folk-Songs of the Southern United States, 35-37--this version contains a brief mention of a ship. W. K. McNeil collected another Kentucky version in 1962 for his Southern Folk Ballads, vol. 1 (Little Rock AR: August House, 1987), 140-41. John Cohen made two recordings of "Pretty Polly" in 1959 by two Kentucky singers, Bill Cornett and Lee Sexton, for his Mountain Music of Kentucky, Smithsonian Folkways, SF 40077 (1959).
(15.) "Certainly the banjo's somber drone and driving rhythmic figures [of B. F. Shelton's recording] receive new life in Bali's sonorous guitar rendition" (Wade, RCA: THE BRISTOL SESSIONS, 1927, 9). "In [Ball's 1941] recording, the guitar style evoked the punctuated qualities of banjo picking: the accompaniment as counterpoint to the melody" (Cohen, High Atmosphere liner notes, 13). The interrelation between Shelton's and Bali's recordings highlights another variable in my examination of recorded performances, that is, the field recording (Ball) and the commercial recording (Shelton). That Ball seemed to model his version of "Pretty Polly" on a commercial recording calls into question the assumption that field recordings preceded and are thus necessarily more authentic than commercial ones. David E. Whisnant writes that "commercial recording of Southern vernacular music seems to have preceded such field recordings" by Lomax and others ("Turning Inward and Outward: Retrospective and Prospective Considerations in the Recording of Vernacular Music in the South," in Sounds of the South: A Report and Selected Papers from a Conference on the Collecting and Collections of Southern Traditional Music [Durham NC: Duke University Press, Southern Folklife Collection, 1991], 165-81, see 165). In addition, Neil V. Rosenberg has commented on this issue in his discussion of early recordings of traditional music: "In the twenties and thirties, American record companies and folksong collectors were sometimes recording the same individuals and groups. Usually neither knew of the other. Frequently the performers were asked for different kinds of song by each. Folksong collectors wanted the older traditional material which they recognized as folksongs, while the record company executives, after the first few years of experimentation, wanted what they considered 'fresh' material. Occasionally a performer's traditional repertoire might be obscure enough to seem 'fresh' to the record company, and it was the presence of recognizable folksongs on hillbilly records--like B. F. Shelton's 'Pretty Polly'--which kindled the interest of folksong scholars in country music. Later, folksong scholars realized that they had recorded new compositions, especially those of Jimmie Rodgers, under the impression that these 'fresh' songs were old folksongs" ("Big Fish, Small Pond: Country Musicians and Their Markets," in Peter Narvaez and Martin Laba, eds., Media Sense: The Folklore-Popular Culture Continuum [Bowling Green OH: Bowling Green State University Popular Press, 1986], 149-66, see 150).
(16.) Almeda Riddle as quoted in Roger D. Abrahams, ed., A Singer and Her Songs: Almeda Riddle's Book o[Ballads (Baton Rouge: Louisiana State University Press, 1970), 122-23.
(17.) Laws, Native American Balladry, 7.
(18.) Personal communication, 7 November 2004.
(19.) Greil Marcus, The Old, Weird America: The World of Bob Dylan's Basement Tapes (New York: Picador, 2001), 179, originally published as Invisible Republic: Bob Dylan's Basement Tapes (New York: Henry Holt, 1997). Reprinted in the liner notes to Dock Boggs Country Blues: Complete Early Recordings (1927-29), John Fahey's Revenant (1997), 26. Marcus also makes the distinction between performances of songs and performances of ballads: "The folk-lyric form came together some time between 1850 and 1875.... [It] sustained the sense that out of the anonymity of the tradition a singer was presenting a distinct and separate account of a unique life. It is this quality--the insistence that the singer is singing his or her own life, as an event, taking place as you listen, its outcome uncertain--that separates the song, from which the singer emerges, from the ballad, into which the singer disappears" (The Old, Weird America, 116).
(20.) Bertrand Harris Bronson, The Ballad as Song (Berkeley: University of California Press, 1969), 131-32. Carl Lindahl argues that this performance aesthetic reflects the traditional ballad singers' innate understanding that their audience would be more drawn in by performances for which they themselves must provide the interpretation. Thus, although ballad singers "possess great powers of introspection, they simply recognize that the strength of their art lies in avoiding interpretation" ("The Oral Aesthetic and the Bicameral Mind," Oral Tradition 6 : 130-36, see 134).
(21.) Riley Baugus, "Pretty Polly," Life of Riley, LORE 0001 (2001). Of course, there are singers who make other performance choices. Bill Ellis describes bluegrass singer Dave Evans's vocal style as "idiosyncratic versions of folksongs.... Taken as denatured texts, his versions seem to tell us little about balladry. But taken as performances--the voice emphasizing, embellishing, sometimes concealing words against a polyphonic instrumental background--Dave Evans's ballads often are disturbing comments on their content and history" ("Reinventing the Anglo-American Ballad: Dave Evans's Performance Style," Southern Folklore 51, no. 3 : 219-40, see 220). Indeed, Evans's approach is remarkable because it stands in such strong relief to the dispassionate style: "Instead of an impassive embodiment of a paradigm, such a performance is a defiant expression of personality" (222).
(22.) Laws, Native American Balladry, 32.
(23.) Thomas G. Burton, in his Some Ballad Folks (ETSU Research Development Committee, 1978), interviewed several ballad singers who felt that the ballads were cautionary and intended to teach young people in particular valuable life lessons.
Buna Hicks: "I'd study out how this was, study it over; and I guess sometimes it might help a body to watch out. Some of 'em that's sung might be a good warning to people sometimes, the lovesongs would. They really, I think, might be to warn somebody if they just take heed and study these songs out. And I think it would be a hope to young folks and get the understandin' of them. It might be a little warning to a body to learn a little more about how to start out." (24)
Hattie Presnell: "Back when my sister was a-talkin' to a boy, they wanted me to sing that 'Pretty Polly' because they thought this boy might not be right fer her, so if she would see and not talk to him. And I told 'em it wasn't none of my business. I didn't sing it to her. But now hit would, some people, I mean hit would give 'em a-somethin' to look at, you know, or somethin' or 'nother like 'at; and might cause a lot of 'em to be careful about goin' off with people." (30) Others spoke of how vivid the frightening images were.
Lena Harmon: "I'd hear my daddy sing one of those murder songs; and aw, it affected my sleep lots of times and then usually would until I'd ask questions to him and I find out a little bit more about it and what he thought about it. Then that would ease my--you, to hear what he thought about. But they scared me, as well as the witch tales he used to tell'd scare me.... Anybody can sing a song, but to sing it and think it over, each person in it, and think on it and get the meaning out of it is another thing.... I picture how this happened and all. If it's a lovesong, or someone was killed, or murdered, or somethin' like that. To me I have a picture in my mind, a very vivid picture of it all. And all my life I could create that scene, and to me it's just somethin' that comes real and handy to me to think on that as being that way.... I believe they are true. Somebody, somewhere, no matter how old a song is, made the song (the words, the tune, and all to it); and of course at that time it really happened that way. And to me, it's as though I'd 'a made it--it's as real to me because I guess I'm just like that." (36, 37)
(24.) Mark Franko and Annette Richards, "Actualizing Absence: The Pastness of Performance," in Mark Franko and Annette Richards, eds., Acting on the Past: Historical Performance Across the Disciplines (Hanover Nit: Wesleyan University Press, 2000), 2. Judith Butler has written extensively on the reiterative nature of performance, particularly with regard to gender as "an identity tenuously constituted in time, instituted in an exterior space through a stylized repetition of acts.... In what senses, then, is gender an act? As in other ritual social dramas, the action of gender requires a performance that is repeated. This repetition is at once a reenactment and reexperiencing of a set of meanings already socially established; and it is the mundane and ritualized form of their legitimation" (Gender Trouble: Feminism and the Subversion o[ Identity [New York: Routledge, 1990], 140, emphasis in original).
(25.) According to Butler, "the possibilities of gender transformation are to be found precisely in the arbitrary relation between such [repeated] acts, in the possibility of a failure to repeat" (Gender Trouble, 141). Thus, in between each iteration a gap exists that allows the possibility for disrupting the conventional performance style. Butler develops her ideas further in Bodies That Matter: On the Discursive Limits of "Sex" (New York: Routledge, 1993); see especially 94-95 and 107-8. "I would suggest that performativity cannot be understood outside of a process of iterability, a regularized and constrained repetition of norms.... This iterability implies that 'performance' is not a singular 'act' or event, but a ritualized production, a ritual reiterated under and through constraint, under and through the force of prohibition and taboo" (95).
(26.) Adrienne Rich, "When We Dead Awaken: Writing as Re-Vision," College English 34 (1972): 18.
(27.) Judith Fetterley, The Resisting Reader: A Feminist Approach to American Fiction (Bloomington: Indiana University Press, 1978), xxiii.
(28.) "Two things are certain: to our folk a ballad is song, not poetry; for us ballads become poems because certain variants (often by sheer chance) measure up to Western European aesthetic standards.... A ballad survives among our folk because it embodies a basic human reaction to a dramatic situation. This reaction is reinterpreted by each person who renders the ballad. As an emotional core it dominates the artistic act, and melody, setting, character, and plot are used only as a means by which to get it across" (Coffin, The British Traditional Ballad, 165). Bronson asks the following: "Question: When is a ballad not a ballad? Answer: When it has no tune.... [S]cholars who disregard the continuous interplay of tune and text in the ballads are guilty of a very one-sided approach to their subject.... Officious nods and becks, theatrical hints of a sub-surface understanding shared between singer and hearer, are an offense to that powerful impersonality which makes good folk-singing so uniquely impressive: they belong rather with the dramatic reading of words-without-notes which makes of the ballads an alien art. That almost marmoreal inviolability of the ballad as traditionally sung subdues insinuations and forbids intimacy. Suggestive inflections of stated meaning, even broadly ironic, find no foothold on this smooth surface.... [T]he persons of balladry should also maintain a directness and simplicity of character incapable of sophistication" (The Traditional Tunes, 1:ix, xi).
(29.) Cook, "Cursed Was She," 216.
(30.) Lily May Ledford, Gems: Lily May Ledford--Rare Concert and Studio Recordings 1968-1983, June Appal Recordings, JA0078D (2000).
(31.) Ledford, Gems, emphasis mine to represent Lily May's delivery. Note that Lily May was from Kentucky; another Kentucky/"Preny Polly" connection?
(32.) Ralph Stanley & Friends, Clinch Mountain Country, Rebel Records, REB-5001 (1997). "Pretty Polly" spent three weeks as number i on the National Bluegrass Survey in 1997. S. Renee Deckert, "Welcome to Patty Loveless's Living Room," review of Patty Loveless's CD Mountain Soul, http://www.popmatters.comlmusic/reviews/l/loveless patty-mountain.shtml, accessed 5 October 2004.
(33.) Despite its commercial origins, a bluegrass performance represents a traditional aesthetic for many contemporary listeners. As Teresa Goddu writes, bluegrass is "self-construct[ed] as a music securely located in the distant past. While bluegrass may appear to be caught in a time warp, it is anything hut static.... Its orientation toward the past is more image than reality, yet it is an image that serves bluegrass well, allowing it to 'get away with' more than mainstream country music can. The nostalgic image of the past that bluegrass propagates can easily accommodate songs of doom and dread" ("Bloody Daggers and Lonesome Graveyards: The Gothic and Country Music," in Cecelia Tichi, ed., Reading Country Music: Steel Guitars, Opry Stars, and Honky-Tonk Bars [Durham NC: Duke University Press, 1998], 45-64, see 57-58).
(34.) I have written elsewhere of the use of quotation as a way to falsely represent the female voice in song. See my "Lost Honor and Torn Veils: A Virgin's Rape in Music," in, Kathleen Coyne Kelly and Marina Leslie, eds., Menacing Virgins: Representing Virginity in the Middle Ages and Renaissance (Newark: University of Delaware Press, 1999), 165-78. According to Laws, "The principal disadvantage of the first-person ballad [is that m]ost ballad stories end in the death of the main character. In a first-person ballad, death may be fearfully anticipated, but it cannot be dramatized, and thus such ballads are not neatly rounded off" (Native American Balladry, 29). Thus, in all the murdered sweetheart ballads the girl will never be able to tell her own story; rather, we will only hear her perhaps through some dialogue as reported by her lover/murderer. The murdered girl can only be present through a change in the performance of a murdered ballad.
(35.) Ruth A. Solie, "The Gendered Self in Schumann's Frauenliebe Songs," in Steven Paul Scher, ed., Music and Text: Critical Inquiries (Cambridge: Cambridge University Press, 1992), 220, emphasis in original.
(36.) Neil V. Rosenberg writes that bluegrass singers from the beginning maintained a traditional singing style: "All [their songs] were sung in a manner which contrasted with popular and classical musics by its impersonality of delivery. The dynamics of the story were not carried over into the dynamics of singing; a humorous novelty item was performed in basically the same way as a serious sad song. This impersonality is a chief characteristic of the Anglo-American folksong performance styles in which most hillbilly musicians were nurtured, a sign of the close links with older traditions exhibited by so many facets of their music" (Bluegrass: A History [Urbana: University of Chicago Press, 1985], 22).
(37.) Judy Collins, Who Knows Where the Time Goes, Elektra Entertainment 74033 (1988; original release date 1968). This recording is available through the iTunes Music Store website.
(38.) Almeda Riddle as quoted in Abrahams, A Singer and Her Songs, 122.
(39.) Hilarie Burhans, Put on the Skillet: Hilarie Burbans Clawhammer Banjo, Make 'em Go Wooo Productions (2003). This recording is available through the iTunes Music Store website.
(40.) Burhans, liner notes. The ellipses are from her notes; I made no omission at the end of her sentence.
(41.) The New Coon Creek Girls, So I'll Ride, Turquoise Records, TR-CD-5075 (1991). The performers are Carmella Ramsey, fiddle, and Vicki Simmons, vocals and banjo. Coincidentally, Carmella Ramsey is Patty Loveless's sister.
(42.) For an introduction to this topic see Joyce H. Cauthen, With Fiddle and Well-Rosined Bow: Old-Time Fiddling in Alabama (Tuscaloosa: University of Alabama Press, 1989), 201-2; Loman D. Cansler, "The Fiddle and Religion," Missouri Folklore Society Journal 13-14 (1991-92): 31-43; and Herbert Halpert, "The Devil, the Fiddle, and Dancing," Fields of Folklore: Essays in Honor of Kenneth S. Goldstein (Bloomington Is: Trickster Press, 1995), 44-54. For material related to the European history of this fiddle/devil association see also Rita Steblin, "Death as a Fiddler: The Study of a Convention in European Art, Literature and Music," Basler Jahrbuch fur historische Musikpraxis: Eine Veroffentlichung der Schola Cantorum Basiliensis an der Musik-Akademie der Stadt Basel 14 (1990): 271-322; and Maximilian Rudwin, The Devil in Legend and Literature (1931; repr. New York: AMS Press, 1970), esp. chap. 20, "The Devil, the World and the Flesh (II): Music, Dance and Drama of Infernal Origin."
(43.) Three of these stories are found in the Violin Times: Aubertine Woodward-Moore, "The Legend of a Violin" (15 December 1899); Anonymous, "The Violin Maker" (three spring issues, 1900); and Marjorie Dillwyn, "His Violin" (October 1902). For more on violin fiction and a summary of these stories see Paula Gillett, Musical Women m England, 1870-1914 (New York: St. Martin's Press, 2000), 92-94.
(44.) "The Twa Sisters" is Child ballad #10 and can be found in Child, English and Scottish Popular Ballads, 18-20. The well-known American version, "The Dreadful Wind and Rain," is a reworking of this ballad. Not surprisingly for an American adaptation, the supernatural elements are somewhat muted, since the fiddle does not reveal the name of the killer. Rather, it will only play one tune called "Wind and Rain." For a discussion of the ballad and the American version see Atkinson, "Magical Corpses and the Discovery of Murder," 185-232, esp. 212-22. See also his "Magical Corpses: Ballads, Intertextuality, and the Discovery of Murder."
(45.) The presence of birds that observe and sometimes reveal a murder in ballads is quite common, particularly in "Lady Isabel and the Elfin Knight" and another ballad, "Young Hunting." Indeed, the name Polly is probably a long-lost remnant of the parrot who often in these ballads witnesses the murders. For a discussion of this trope see David Atkinson, "Motivation, Gender, and Talking Birds," in his The English Traditional Ballad, 146-84, esp. 171-80.
(46.) Traditional performers (e.g., Estil C. Ball, Dock Boggs, and B. F. Shelton) also alter the lengths of phrases but often in a seemingly random way. However, there is an astute and considered quality to the manipulation of phrase lengths in the New Coon Creek Girls' version, with this traditional feature serving a powerful expressive function.
(47.) Fetterley, The Resisting Reader, xxiii-xxiv.
(48.) Fenerley, The Resisting Reader, xxiii, emphasis mine.
(49.) k. d. lang's performance of fifties hit "Johnnie Get Angry" is another example of a powerful resisting performance that operates within the sound context of the original. See k. d. lang, Harvest of Seven Years (Cropped and Chronicled) (Burbank CA: Warner Brothers, VHS 1991, DVD 2002).
(50.) Edith Randam Rogers, The Perilous Hunt: Symbols in Hispanic and European Balladry (Lexington: University Press of Kentucky, 1980), 148, as quoted in William Bernard McCarthy, The Ballad Matrix: Personality, Milieu, and the Oral Tradition (Bloomington: Indiana University Press, 1990), 132.
TABLE 1. Number of Text * Beats ** Fiddle Stanza 1 I used to be a rounder, I've been around this town 8+6 Banjo only I used to be a rounder, I've been around this town 8 I courted Pretty Polly, I've been all around. 8+10 Stanza 2 Oh where's Pretty Polly, over yonder she stands 8+6 First fiddle entrance; appoggiaturas; low Oh where's Pretty Polly, over yonder she stands 12 High With diamonds on her fingers and her lily-white hands. 8+10 Polly melody-a fifth Stanza 3 Oh Polly, Pretty Polly, come go along with me 8+6 Double-stops; low Oh Polly, Pretty Polly, come go along with me 12 High Before we get married some pleasures you'll see. 8+10 Polly melody-a fifth Stanza 4 I led her over mountains and valleys so deep 8+6 Double-stops; low I led her over mountains and valleys so deep 8 Low But then Pretty Polly, she began to weep. 8+10 Abbreviated Polly melody, ends low--a twelfth Stanza 5 Oh Willie, oh Willie, 8+6 Melody similar I'm afraid of your ways to 2 but high Oh Willie, kind Willie, I'm afraid of your ways 16 High; agitated I'm afraid you are 8+10 Polly melody-a goin' to lead me astray. fifth Stanza 6 Oh Polly, Pretty Polly, your guess is about right 8+6 Double-stops; high Oh Polly, Pretty Polly, your guess is about right 16 Low I dug on your grave the 8+36 Extended agitated biggest part of last night. melody-two octaves Stanza 7 Oh Willie, oh Willie, 8+6 Melody of S; please spare me my life double-stops; high dissonant; Oh Willie, kind Willie, please spare me my life 12 High; syncopated But into her bosom 8+6 High; short; I plunged the fatal knife. syncopated; dissonant Stanza 8 She fell to the ground and the blood it did flow 8+6 Diapente melody; low She fell to the ground 8 High but within and the blood it did flow a fifth I left Pretty Polly for the birds to weep and moan. 8+10 High trills of birds Stanza 9 And now to the devil a 8+6 Double-stops in debt I must pay fourths; low And now to the devil 12 Double-stops in a debt I must pay seconds; low For killing Pretty 8+10 Constant, syncopated Polly and running away. double-stops; Polly melody somewhat curtailed; gives way to the main melody, which then slips into the flat7 region and harmonics * Words in italics indicate where phrase extentions occur. ** Numbers in italics represent the responses to phrases and the interludes between stanzas. Source: Created by the author.