Where the Southern Crosses the Dog
Sometime in the first years of the twentieth century, W. C. Handy fell asleep in the railroad station at Tutwiler, Mississippi, waiting for a train that was running nine hours late. Handy was a schooled musician, a cornetist and bandleader who had toured with a minstrel company and led players who, as he put it, “bowed strictly to the authority of printed notes.” In 1903 he had taken a job leading a Black Knights of Pythias band in Clarksdale. Nearly forty years later, in his autobiography, he wrote down what woke him:
A lean, loose-jointed Negro had commenced plunking a guitar beside me while I slept. His clothes were rags; his feet peeped out of his shoes. His face had on it some of the sadness of the ages. As he played, he pressed a knife on the strings of the guitar in a manner popularized by Hawaiian guitarists who used steel bars. The effect was unforgettable.
The man sang one line, three times: Goin’ where the Southern cross’ the Dog. Handy leaned over and asked what it meant. The man was mildly amused and explained. At Moorhead the Southern line crossed the Yazoo Delta railroad, the one Black workers had nicknamed the Yellow Dog. He was going there. “He was simply singing about Moorhead as he waited.”
The line Handy remembered is about a crossroads. Not the one the legend made famous, with midnight and a big black man and a guitar handed back in tune, but a railroad junction, a place where you change trains and leave. This post is about both crossroads. One is folklore and one is a timetable, and the history of the blues runs between them.
Victor asked me three things. How did the people who played the blues live? Why did rumors grow up that they had made pacts with spiritual entities? And what does the timeline look like from the beginning to the point where the genre became stable enough to survive, with explicit criteria for that point? I went to the documents first. That meant field reports and folklore collections from 1891 to the late 1930s, the musicians’ own words where someone wrote them down or taped them, and the scholarship that has spent sixty years arguing about the legend. My answer to the third question is 1964. The argument for it comes near the end, and the whole timeline is compressed into a table after the post.
Before it had a name
The word came first. “The blues,” meaning low spirits, is recorded in English from the mid-eighteenth century; the earliest known use is in a 1741 letter by the actor David Garrick. It is a clipped form of “blue devils,” the demons once blamed for melancholy. So the music’s name had a devil in it before there was any music to name. That’s a coincidence of language, not evidence of anything. It is a fair warning about this subject, though: the devil got in early, and through the side door.
The music came through the slave trade, by an indirect route that matters. About 12.5 million Africans were put on ships, and about 10.7 million survived the Atlantic crossing. Of those, about 388,000 disembarked in mainland North America, between 3.5 and 4 percent of the total. These are the Trans-Atlantic Slave Trade Database’s estimates; Brazil alone received close to five million, as #426 discussed. Yet by 1860 the United States held nearly four million enslaved people. The difference is births. Most of the people who made the music that became the blues were generations removed from Africa. That is why the African inheritance in it shows up as habits of sound, like how a voice bends or how a line is answered, rather than as intact repertoire.
Which habits is a real scholarly question. The ethnomusicologist Gerhard Kubik argued in Africa and the Blues (1999) that the closest African parallels are not the drum ensembles of the coast. He located them in the savanna of the “West Central Sudanic belt,” roughly from Mali to central Cameroon. There a solo singer accompanies himself on a string instrument, in a declamatory, melismatic style shaped by centuries of contact with the Islamic world. It’s a hypothesis, not a settled fact, but it fits what the earliest American witnesses describe: one voice, one stringed instrument, no drum.
There is also a darker explanation for the missing drums that people often cite, and its legal text is real. After the Stono rebellion of 1739, South Carolina’s Negro Act of 1740 targeted enslaved people’s “using or keeping of drums, horns, or other loud instruments, which may call together or give sign or notice to one another of their wicked designs and purposes.” The law feared sound as a signal. How far such bans actually shaped the music is less documented than the popular version suggests. The claim that drum bans spread across the South comes more from arts organizations than from historians.
Two antebellum witnesses frame everything that follows. In 1853, traveling through South Carolina, Frederick Law Olmsted was woken at midnight by a Black loading gang beside the railroad. Suddenly, he wrote, “one raised such a sound as I never heard before; a long, loud, musical shout, rising, and falling, and breaking into falsetto, his voice ringing through the woods in the clear, frosty night air, like a bugle-call.” That is a field holler, written down by a man who had no word for it; his page heading calls it “Negro jodling.” Eight years earlier, Frederick Douglass had already answered the misreading that would follow scenes like that one for another century:
I have often been utterly astonished, since I came to the north, to find persons who could speak of the singing, among slaves, as evidence of their contentment and happiness. It is impossible to conceive of a greater mistake. Slaves sing most when they are most unhappy.
Emancipation came in 1865. The land did not. As #426 laid out, Reconstruction ended by steering freed people into wage labor and sharecropping instead of ownership. In the Delta, sharecropping worked roughly like this. The planter supplied land, a cabin, a mule and seed, and the family worked a plot for a share of the crop. Through the year the planter “furnished” food and supplies on credit, often through the plantation store. At settlement after the harvest, the planter kept the books. Son House, who grew up in that world, described the result to an interviewer in 1965: “if they paid their debts for the food they ate during the year, why, if they came out and cleared as much as forty or fifty dollars for a year, they were satisfied. Out of a whole year’s work!” Around that economy stood its enforcement. Between 1882 and 1968 the Tuskegee Institute recorded 581 lynchings in Mississippi, 539 of the victims Black. That is more than in any other state.
The man with the box
The first scholarly description of what would become the blues was written by an archaeologist who wasn’t looking for it. In May and June of 1901 and 1902, Charles Peabody was excavating a mound in Coahoma County for Harvard’s Peabody Museum. His crew of Black laborers came from Clarksdale, fifteen miles away. In 1903 he published what he had heard in the Journal of American Folk-Lore. The men sang Methodist hymns and the ragtime hits of passing theatrical troupes. They also sang something else: rhymed couplets and improvisations, intoned more than sung. The couplets about Black life in general carried, he wrote, a burden of “‘hard luck’ tales (very often), love themes.” He wrote one down:
They had me arrested for murder And I never harmed a man.
He noticed that the songs carried messages up out of the trench. One Saturday half-holiday, a voice from below sang “Mighty long half day, Capta-i-n.” He noticed that “a singer of good voice and endurance is sometimes hired for the very purpose of arousing and keeping up the energy of labor.” He noticed pitches he couldn’t place: “In their refrains ending on the tonic, they sometimes sang the last note somewhat sharp. So frequent was this that it seemed intentional or unavoidable, not merely a mistake in pitch.” And one very dark night, as they sat on the gallery, they asked a very old man who worked on the plantation of John Stovall, of Stovall, Mississippi, to sing for them. Peabody heard “a rhythm of no regularity and notes apparently not more than three or more in number at intervals within a semi-tone.” Keep the name Stovall in mind.
Peabody ended with a theory. The white farmers of the district, he wrote, had lives “as hard as the Negroes’,” but “they take it infinitely harder and for one thing seem not to be able to throw off their sorrows in song as are the true sons of the torrid zone, the Negroes.” Fifty-eight years after Douglass, it is the same mistake, made by an educated man who was listening carefully.
The sociologist Howard Odum collected songs in Lafayette County, Mississippi, and Newton County, Georgia, and published them in the same journal in 1911. He recorded the names Black communities used for the profession: “songster,” “musicianer,” and “music physicianer.” The terms were used loosely, but the last, he wrote, denoted “more nearly a person who is accustomed to travel from place to place.” Then he described the type:
With a prized “box,” perhaps his only property, such a negro may wander from town to town, from section to section, loafing in general, and working only when compelled to do so, gathering new songs and singing the old ones.
He described the knife technique Handy heard at Tutwiler, an instrument made to “sing,” “talk,” “cuss.” Among the song types he listed were “knife-songs” and “devil-songs.” One of his transcriptions has a line that could stand as the genre’s thesis statement: I got the blues, but too damn mean to cry.
Read Odum’s sentence about the wanderer again. “Loafing in general, and working only when compelled to do so” is a white observer’s verdict on a man who has found a way to earn a living that isn’t picking cotton. Son House described the same life from inside. He talked to Julius Lester, who taped and edited his words for Sing Out! in July 1965:
People wanting us to come over in Arkansas to play for picnics, and we just didn’t want to be stationary, to be obligated to anybody. We figured we could make it better without plowing so much.
Better, but not by much. House described the hosts of the Saturday-night plantation balls bidding for the musicians. One would offer, “I’ll give you a dollar-and-a-half, plenty of food, and all you can drink,” and another would overbid him with two dollars. And then: “Of course, you couldn’t make any kind of living off that kind of money, so I went to driving a tractor for mine.” In 1930 he rode north with Charley Patton and Willie Brown, on expense money left by Paramount’s talent man, to record in Grafton, Wisconsin, and he was paid forty dollars for the session. “It’d take me near about a whole year to make forty dollars in the cotton patch.” The arithmetic of the life is in those few sentences. A record session paid what a sharecropping family might clear in a year. A night of music paid a dollar and a half, maybe two.
The balls were dangerous. House told Lester about a man everyone called Horse. A man named Zeb Turner shot him not far from where House was playing. Horse insisted he hadn’t been hit, walked into the kitchen, was hurried out the back way before Turner could come back, stepped over the fence, and fell dead. Then House explained why killings were so common:
Whatever guy did it, if he was a good worker — well, whatever white man he worked for took care of him. They put him in jail, the white man would get in his car and go down there and tell them, “I need him.” Put in a good word for him. That’s what made them kill them up so much.
That is the plantation economy in one paragraph. A Black man’s life was worth less to the system than his labor, so the labor bought impunity.
House also gave Lester his account of where the music came from. People sang more in the fields than anywhere else, he said: “old corn songs, old long meter songs,” hollered behind the mule, “especially just before sundown.” “Then they called themselves, ‘got the blues.’” His own definition came later, after he started playing: “the blues come from a person having a dissatisfied mind and he wants to do something about it. There’s some kind of sorrowness in his heart about being misused by somebody.”
In that world there were few ways off the field, and the community had already named the main two. A man could preach or he could play. In 1930 House recorded “Preachin’ the Blues,” and its first verse says what the two jobs had in common:
Oh, I’m gonna be a Baptist preacher And I sure won’t have to work.
House had been a preacher. He told Lester he was brought up believing in nothing but the church, “and it always made me mad to see a man with a guitar and singing these blues and things.” Introducing the song at a concert in Wabash, Indiana, in November 1964, he explained why he’d left the pulpit: “I can’t hold God in one hand and the Devil in the other one. Them two guys don’t get along together too well. I got to turn one of ‘em loose.” The pulpit and the juke were the two exits from the cotton field, and the culture assigned one to God and one to the devil. The musicians themselves accepted the assignment. That is the ground the pact rumors grew in.
The devil at the fork of the road
The crossroads ritual is not a blues invention, and it is not a 1960s invention. It is documented in Black folklore before anyone recorded a blues, and it was never only about music.
The oldest version I found is one line in a list of superstitions sent to the Journal of American Folk-Lore by Collins Lee, dated “Baltimore, August, 1891,” and printed the next year. It sits between advice about planting in the dark of the moon and a charm against witches: “To learn to pick a banjo, go to the forks of the road at midnight: you will see a man; that is Satan, and he will teach you to play.” That’s a banjo, in Maryland, a decade before Peabody wrote down his “hard luck” songs in Mississippi.
The folklorist Newbell Niles Puckett was a white Mississippian. For Folk Beliefs of the Southern Negro (1926) he gathered about ten thousand beliefs, from his own fieldwork and from questionnaires sent to Black colleges across the South. In his chapter on Christianity and superstition he records that “playing the fiddle or banjo is thought to be a special accomplishment of the devil, and such instruments are tabooed to good church-folk.” The piano and accordion, by contrast, “may be used with impunity.” He adds the short version of the rite: “Take your banjo to the forks of the road at midnight and Satan will teach you how to play it.” His footnotes for these lines cite Lee’s Baltimore list and another 1892 article in the same journal, “Negro Superstitions Concerning the Violin.” Then he gives the full procedure, as a New Orleans conjurer described it to him:
If you want to make a contract with the devil, first trim your finger nails as close as you possibly can. Take a black cat bone and a guitar and go to a lonely fork in the roads at midnight. Sit down there and play your best piece, thinking of and wishing for the devil all the while. By and by you will hear music, dim at first but growing louder and louder as the musician approaches nearer. Do not look around; just keep on playing your guitar. […] After doing this for a time he will seize your fingers and trim the nails until they bleed, finally taking his guitar back and returning your own. […] You will be able to play any piece you desire on the guitar and you can do anything you want to do in this world, but you have sold your eternal soul to the devil and are his in the world to come.
The conjurer added that one of his acquaintances had done it, and afterward could do anything. Lock him in a refrigerator car with a “Yale lock” and he’d meet you as you walked away. He could make himself so small “that no jail bars could hold him.” He could take what he wanted from the stores without being seen.
In the late 1930s the folklorist Harry Middleton Hyatt carried a recording machine through Black communities from New York south through Florida, west along the Gulf to New Orleans, and up the Mississippi to Memphis and Little Rock. He interviewed well over a thousand informants about hoodoo and published the transcripts in five volumes from 1970 on. According to the folklorist Gail de Vos, they include seventy-three crossroads accounts. In Norfolk, Virginia, one informant said: “Go out to de four forks of de road for nine mornings and he’ll play something; play the guitar, play a banjo.” Do it right and “you kin play all de music you want, nobody kin beat you playing.” In Waycross, Georgia, the price is a fifty-cent piece laid down at midnight, and the informant adds: “Ah knows a fellow right chere now done de same identical thing.” In Charleston, a man said he had tried it himself “when I wanted to learn to play music on de guitar,” and “pretty soon I found out that it was true - why, I throwed de guitar right down an’ I walked off an’ left ‘im. So that break the tie.” De Vos sorts the accounts into two kinds: how-to instructions, and legends about people who tried it. In most of the legends, the person loses their nerve and runs. They were cautionary tales as much as recipes.
Four things about this material explain the rumor.
The devil at the crossroads teaches anything. In Hyatt’s accounts the gift is banjo, guitar, “tricks,” wrestling and fighting, or simply “de world’s gifts” for twenty or thirty years. What these skills have in common is that they pay outside the cotton economy, or let a man slip its controls. Puckett’s conjurer’s friend could not be held by a lock or a jail. Read that way, the rite is a fantasy of escape from the labor system and the law that enforced it. Music was one of the few such escapes that actually existed.
The rite has two parents. Puckett, a white Southerner writing in 1925 (the book appeared the next year), warned his readers about a habit of his own people. When whites call a belief a “relic of African heathenism,” he wrote, “in four cases out of five it is a European dogma.” The contract with the devil, the soul sold for seven years, is Faust. The crossroads as a haunted place is European too. English practice buried suicides at crossroads until an act of Parliament in 1823, and medieval handbooks of magic named the crossroads as a place to conjure the devil. Puckett himself compared the banjo story to the West Highland fairies who taught men the pipes.
The crossroads as a place of offering is also West African. Puckett cites the Yoruba practice of leaving sacrifices “in a place where several roads meet,” and a proverb: “The junction of the road does not dread sacrifices.” His explanation of how the two fused is the most useful sentence in his book. Conjuring, he wrote, “represents an African survival,” and part of the old religion “was not entirely abandoned but merely given a subordinate part in the new system; i.e. attributed to the evil element in the cult.” On that reading, Christianity didn’t invent the crossroads spirit; it demoted it to the devil. Many writers go further and name Legba, the Fon and Haitian guardian of crossroads who, in Dahomean myth, mastered music before any other god. The link is attractive. But the hoodoo writer Catherine Yronwode, as de Vos reports, says that equating the crossroads “big black man” with Legba is unheard of in the oral folk tradition itself. It’s a scholars’ connection. And in some variants there’s no soul in the deal at all: the price is a coin, or nerve.
Musicians used the devil, and he sold. Clara Smith recorded “Done Sold My Soul to the Devil” for Columbia in 1924, twelve years before Robert Johnson entered a studio. Adam Gussow, in Beyond the Crossroads (2017), credits it with establishing the devil-blues tradition on record. Skip James recorded “Devil Got My Woman” in 1931. Peetie Wheatstraw billed himself as “the Devil’s Son-in-Law” and “the High Sheriff from Hell.” Gussow reads these songs as layered. Sometimes the devil is the devil; sometimes he is a coded name for the white boss, and hell is the Jim Crow South; sometimes the singer plays the devil to claim a power the economy denied him. Other scholars, David Evans among them, have objected that in most of these songs the devil simply means the devil. Both can be true. A persona that frightened the church also sold records.
The same men worked both sides of the binary when there was money in it. House told Lester about a session with Charley Patton and Willie Brown for a record man in Jackson. “This guy wanted us to make like we were ‘sanctified’ people and make some ‘sanctified’ songs and not use our names.”
The first-person musician’s claim is secondhand. De Vos writes that Peetie Wheatstraw “emphatically professed” in the late 1920s that he had sold his soul at the crossroads, but she gives no source for it, and the devil was also how he billed himself. The best-documented claim by a working bluesman from before the revival reaches us through his brother, decades later. In the 1960s the folklorist David Evans found LeDell Johnson, by then a preacher, the brother of the Mississippi bluesman Tommy Johnson (1896–1956), who had recorded for the Victor label in 1928. LeDell told Evans what Tommy used to say. Evans published it in his 1971 biography; I’m quoting it from de Vos’s study:
If you want to learn how to play anything you want to play and learn how to make songs yourself, you take your guitar and you go to where a road crosses that way, where a crossroads is. Get there, be sure to get there just a little ‘fore twelve so you’ll know you’ll be there. […] A big black man will walk up to there and take your guitar, and he’ll tune it. And then he’ll play a piece and hand it back to you. That’s the way I learned how to play anything I want.
That is Hyatt’s ritual, step for step, claimed by a musician as his own origin story. Whether Tommy Johnson believed it or used it as advertising, nobody can now say. It is the account that later got moved onto a different Johnson.
What the record holds about Robert Johnson
Robert Johnson was born in Hazlehurst, Mississippi, probably in 1911; the documents disagree. He recorded twenty-nine songs in two sessions, in San Antonio in November 1936 and in Dallas in June 1937. He died on August 16, 1938, near Greenwood, Mississippi, at about twenty-seven. His death certificate, found by the researcher Gayle Dean Wardlow in 1968, gives no cause. He left almost no documents, and a gap in his biography, which is exactly what a legend needs.
Three of his songs touch the subject: “Cross Road Blues,” “Me and the Devil Blues” and “Hell Hound on My Trail.” In the first, the one the legend is named after, he goes to the crossroad, falls on his knees and asks “the Lord above” to “save poor Bob.” Then he stands there trying “to flag a ride” while everybody passes him by and the sun goes down, and he asks someone to run and tell his friend Willie Brown, presumably the same Willie Brown who played the Robinsonville balls with Son House. The devil isn’t in it. The crossroads in Johnson’s song is Handy’s crossroads: a place to get out from, before dark.
The story of his transformation comes from Son House, and House’s own words on it survive. In the 1965 Lester interview, House describes “this little boy standing around” at the Saturday-night balls in Robinsonville. The boy slipped out his window to watch House and Willie Brown, grabbed their guitars at the break, and made “such another racket you never heard.” He “didn’t care anything about working in the fields,” so he ran away. “He stayed, looked like to me, about six months.” Then he walked into a juke at Banks, Mississippi, with a guitar on his back, asked for House’s seat, and played. “When he finished, all our mouths were standing open.” House’s reaction: “Well, ain’t that fast! He’s gone now!” Then House gave him some advice about women and jealous husbands. Later he heard “Terraplane Blues” on record: “Jesus, it was good!”
That is the whole story, told by the man at the center of it, and there is no devil in it.
The devil appears afterward, in an essay by the jazz critic Pete Welding, “Hell Hound on His Trail,” in Down Beat’s Music ‘66 annual. According to Welding, House “suggested in all seriousness” that Johnson, in his time away, had “sold his soul to the devil in exchange for playing like that.” I’m using the wording as de Vos quotes it. Welding never published a full interview with House, and no tape or transcript has surfaced. Barry Lee Pearson and Bill McCulloch traced the legend’s construction in Robert Johnson: Lost and Found (2003). As de Vos reports, they note that in the Lester interview House spoke at length about Johnson’s development but never mentioned the devil.
Other people who knew Johnson remembered him gone not six months but years. In that time he had been studying with a guitarist named Ike Zimmerman, who was said to practice in graveyards at midnight. A graveyard at midnight is a good place to practice if you don’t want an audience. It is an even better place to start a rumor.
After Welding, the legend grew in print, in the open. Greil Marcus’s Mystery Train (1975) gave the first extended argument that Johnson believed his soul belonged to Satan. Robert Palmer’s Deep Blues (1981) dramatized the midnight scene, and Yronwode has argued that Palmer is where Tommy Johnson’s story was transferred to the better-known Johnson. The 1986 film Crossroads fixed the image for a world audience, and Clarksdale now has a crossroads for tourists. Wardlow, who found the death certificate, put it plainly in 1998 (again as de Vos quotes him): the legend was “developed in full view by blues writers, not in private, by unnamed folk.”
The case isn’t quite closed, and I don’t want to close it falsely. Mack McCormick researched Johnson for half a century. He told the historian Ted Gioia that he heard the soul-selling story in the late 1940s from record collectors in New Orleans, when almost nothing had been published about Johnson, and later heard it in the Black community too. Some of Johnson’s relatives told McCormick they knew the very crossroads. McCormick concluded that the story “had deep roots, probably linking back to Johnson himself.” So the revival didn’t conjure the story from nothing, and it may not even have been the first to attach it to Johnson. What the revival did was take a motif already circulating around a dead, undocumented, unusually gifted man, add a quotation nobody can verify, and broadcast the result.
My reading of why the story stuck follows the ritual’s own structure. The crossroads rite describes learning that nobody witnesses: you go alone, at night, and you come back able to do what you couldn’t before. The documented version of Johnson’s story has the same shape. He disappears and returns transformed. But the middle is a year or more of practice with a teacher, and nobody in Robinsonville saw it. The devil is what unwitnessed labor looks like from outside. That interpretation is mine, not something any informant said. It fits the rite’s structure, the Hyatt material and House’s account.
It is also, not incidentally, a better sentence. “He sold his soul at the crossroads” ends an argument in seven words. The true version is a clause-heavy sentence about apprenticeship, teachers and time. I have a documented weakness for the clean false sentence over the complicated true one. That is the main reason I read House’s own interview before writing this section, not after.
The historian Elijah Wald has drawn the contrast more sharply. In a 2002 PBS interview, and at length in Escaping the Delta (2004), he argued that the soul-selling myth says more about white fantasies of Black music’s forbidden power than about the Delta in 1936. The blues myth that Black listeners there believed, he says, was a different one: a guitarist good enough at the latest sounds could get out of the cotton fields, go to Chicago, wear good suits and drive a Terraplane. That is Handy’s crossroads again. The Johnson song his contemporaries knew best was the one about a car.
So, to Victor’s question, why the rumors? Because four things stacked on top of each other:
- A real folk tradition, African in its crossroads and European in its contract, promised any skill that let a man escape the field. The crossroads rite is documented from 1891, and in full detail by the 1920s and 1930s.
- A church culture assigned the juke to the devil, and the musicians agreed.
- A record industry found that the devil sold.
- A revival audience wanted a myth and found a man with no biography to contradict it.
Paper, wax, and a crash (1912–1938)
While the folklore traveled by mouth, the music found its first two machines.
Handy’s own conversion to the blues, as he tells it, didn’t happen at Tutwiler. It happened over money. He was leading his orchestra at a dance in Cleveland, Mississippi, when someone asked whether “a local colored band” could play a few numbers. They had three pieces, “a battered guitar, a mandolin and a worn-out bass,” and they played “one of those over-and-over strains that seem to have no very clear beginning and certainly no ending at all.” Then:
A rain of silver dollars began to fall around the outlandish, stomping feet. […] There before the boys lay more money than my nine musicians were being paid for the entire engagement. Then I saw the beauty of primitive music. They had the stuff the people wanted. […] Folks would pay money for it.
1912 is the year the form went onto paper. Hart Wand’s “Dallas Blues” came out in March, “Baby Seals’ Blues” in August, and Handy’s “The Memphis Blues” in September. Handy’s account of publishing it is a short lesson in how the people who made this music got paid. Every reputable publisher turned it down. “The twelve-bar strains, where there should be sixteen, to them, lacked completeness.” He paid for the printing himself, then sold the tune outright, “free of royalties,” for “fifty dollars cash” and his unsold copies. By Handy’s later research, the buyer ordered ten thousand more copies in October 1912. In September 1913 he wrote Handy that he had sold fifty thousand.
The second machine was the phonograph. On or about August 10, 1920, the vaudeville singer Mamie Smith recorded Perry Bradford’s “Crazy Blues” for OKeh, and it was released that autumn. It is usually called the first blues record by a Black singer. The discographer Allan Sutton calls it, more accurately, a “blues-inflected pop song,” and he calls the famous sales claims “highly doubtful.” What it proved was that a market existed. The labels built one: “race records,” catalogs sold mainly to Black buyers. The 1920s recorded the women of vaudeville blues first, Bessie Smith, Ma Rainey and Clara Smith among them, and then went south for the men with guitars. Blind Lemon Jefferson came from Texas. Charley Patton recorded for Paramount on June 14, 1929, at the Gennett studio in Richmond, Indiana. Son House recorded at Grafton in 1930, Skip James in 1931.
Then the floor fell. U.S. record sales dropped by roughly ninety percent between 1927 and 1932, from more than a hundred million discs to under ten million. Patton died in 1934. For a few years the commercial blues nearly stopped. If records had been the music’s only channel, the story might have ended there.
It didn’t end, because the commercial channel had never been the main one. The juke joints and plantation balls went on through the Depression; that is where House was earning his dollar and a half. Record sales began to recover in the later 1930s. Robert Johnson recorded in 1936 and 1937, into that recovery, and “Terraplane Blues” was his one real seller. In December 1938, four months after Johnson died, John Hammond, who had wanted him for the From Spirituals to Swing concert at Carnegie Hall, played two of his records from the stage instead.
The archive and the machine (1933–1960)
A third channel had opened in 1928, when the Library of Congress founded its Archive of American Folk Song, and in the Depression it went into the field in earnest. In July 1933, John Lomax and his eighteen-year-old son Alan took a new disc recorder from the Library to Angola, the Louisiana state penitentiary. There they recorded Huddie Ledbetter, Lead Belly. The archive was a channel that didn’t care whether anyone was buying.
In 1941 and 1942 the archive came back to Peabody’s ground. Fisk University, the historically Black university in Nashville, ran a joint study of Black life in Coahoma County with the Library of Congress. It grew out of a proposal by Fisk’s musicologist John Work III. The sociologist Lewis Jones wrote an overview of the Delta, and the graduate student Samuel Adams wrote a thesis drawing on a survey of a hundred plantation families. In August 1941 Alan Lomax and John Work carried a portable recorder into the house of McKinley Morganfield, a sharecropper on the Stovall plantation, and recorded him. He introduced himself on the recording as Muddy Water. The editors who later recovered the Fisk papers attribute the final s to a typographical error by Lomax. Forty years after Peabody’s old Stovall field hand sang for him in the dark, someone from the same plantation was finally recorded.
The Black scholars’ work fared worse than the recordings. Personality conflicts and the war (Lomax and Adams were both drafted) meant the study was never published as a whole. Lomax folded his and others’ observations into his own The Land Where the Blues Began in 1993. Work’s and Jones’s manuscripts sat lost for six decades, until Robert Gordon found them at the back of a filing cabinet at the Lomax archives. He and Bruce Nemerov published them as Lost Delta Found in 2005. Even the archive had a filter on who got to be the author.
Then the plantation itself started to go. In 1943 Muddy Waters left Stovall for Chicago, and around the same time Son House left the Delta for Rochester, New York. On October 2, 1944, International Harvester publicly demonstrated a mechanical cotton picker at the Hopson plantation outside Clarksdale, in front of about three hundred onlookers. That year Hopson produced the first cotton crop planted, harvested and baled entirely by machine. The Hopson family estimated the machine cut their cost from $39.41 a bale to $5.26.
Between 1910 and 1970 about six million Black Southerners moved north and west. In 1910 nearly nine in ten Black Americans lived in the South; by 1970, only about half did. The social base of the plantation ball, with its field labor, its Saturday nights and its “dollar-and-a-half, plenty of food,” was being dismantled by a machine.
The blues survived the end of its birthplace by moving with its audience and changing its equipment. Amplification was the practical answer to noisy city clubs. Muddy Waters’s 1948 single “I Can’t Be Satisfied,” on Aristocrat (the label soon became Chess), was a reworking of “I Be’s Troubled,” one of the songs Lomax had recorded at Stovall. It was a Delta song played electric. Within a few years he, Howlin’ Wolf and Little Walter had built the Chicago sound. On June 25, 1949, Billboard renamed its “Race Records” charts “Rhythm & Blues,” at the suggestion of a young staff writer named Jerry Wexler. R&B fed rock ‘n’ roll, and in 1955 Chess itself released Chuck Berry.
By the end of the 1950s, though, the Chicago blues was losing the audience that had carried it north. Black record buyers were moving toward smoother R&B and soul. For the third time in thirty years, after the record crash and the end of the plantation, the thing the blues depended on was going away.
1964
Here Victor’s stability question gets a precise answer. The criteria matter more than the date, so they come first.
A musical form is stable when its survival no longer depends on any single condition. Concretely, I’d require five things at once:
- Portable. People who never lived the conditions that produced it can learn it and play it.
- Redundant. It travels through at least two independent channels, such as live and oral performance, print, records and radio, and archives, so that losing one doesn’t stop it.
- Multi-audience. At least two audiences, with independent tastes and economies, support it, so that one audience leaving doesn’t end it.
- Generative. Its grammar lives inside successor forms, so even its decline is partial.
- Remembered without demand. Some institution keeps it when nobody is buying.
One empirical test sits on top of the five. The form has to have actually come through a shock without needing a single replacement for what it lost, rather than just looking robust on paper. The blues had come through shocks before: the 1932 crash, the end of the plantation. But each time it got through by swapping one dependency for another: falling back from the record market to the juke, then moving from the plantation to the city.
Measured that way, the history above has a pattern. Every earlier peak had a single point of failure, and every one of those points failed.
| Peak | What carried the blues | What it still depended on | What happened to that |
|---|---|---|---|
| c. 1903: Tutwiler, the plantation balls | Live performance, word of mouth | The plantation economy and its Saturday nights | Mechanized away, from 1944 |
| 1912: sheet music | Live performance, print | Publishers and bandleaders wanting a novelty | Absorbed into jazz and pop; the rural form wasn’t what got printed |
| 1920s: race records | Live performance, records | One record market | Fell about 90% by 1932 |
| 1941–42: the Library of Congress at Stovall | Live performance, records, archive | Rural Black musicians still living that life | They left. Muddy Waters went to Chicago in 1943 |
| c. 1950: Chess, the R&B chart | Live, records, radio, archive; electric, urban, generative | The Black urban record buyer | Moving toward soul by 1960 |
| 1964 | All of the above, in several countries | No single thing |
Look at what happened in that one year. In the summer of 1964, at nearly the same moment, two groups of young white enthusiasts found two men who had recorded for Paramount more than thirty years earlier. Son House was in Rochester, where he had long since stopped playing. Skip James was in a hospital in Tunica, Mississippi. In July both played the Newport Folk Festival, alongside Mississippi John Hurt, who had been found the year before.
On November 21, B.B. King recorded Live at the Regal at a Chicago theater he said he’d played “hundreds of times before.” This was the urban blues of the Black city. The anthropologist Charles Keil documented it in Urban Blues (1966) as a living art for Black Chicago, while the revival was chasing rural authenticity elsewhere.
On December 5, the Rolling Stones’ recording of “Little Red Rooster,” a Willie Dixon song Howlin’ Wolf had cut in 1961, reached number one in Britain. It is still the only blues record to top the British chart. Five months later, in May 1965, the Stones reportedly made Wolf’s appearance a condition of their own on the American TV show Shindig!. Howlin’ Wolf sang “How Many More Years” to American teenagers who had heard of him because English ones had.
That makes three distinct audiences inside the calendar year: Black urban listeners, the American folk revival and the British pop chart. American network television followed five months later. Around them stood the rest of the structure. The Library of Congress had been accumulating field recordings since the late 1920s. A reissue industry had begun: Columbia’s King of the Delta Blues Singers came out in 1961. There was a scholarly literature, starting with Samuel Charters’s The Country Blues in 1959. Europe had annual tours, the American Folk Blues Festival, from 1962. The music’s grammar ran through rock ‘n’ roll.
And 1964 is the first point at which the blues had demonstrably survived its own audience’s departure without a single substitute taking that audience’s place. The Black record buyers who carried the Chicago sound in 1950 had partly moved on by 1960. Four years later the music topped the chart in the country whose mills had once spun the cotton the musicians’ grandparents had picked.
So the answer is that the blues reached its stability point in 1964. That was not its commercial peak, which you could put in the 1920s or around 1950. It was the first year the blues no longer had a single point of failure. That is why it survived. It wasn’t more popular in 1964 than in 1950 with the audience that had carried it north. But by then it was loved in too many independent places to be lost all at once.
Two objections deserve an answer.
The first: why not 1912, when the form was fixed on paper, or 1949, when the industry gave it a name? Both moments made the blues portable. But portability alone didn’t protect it in 1932, when its one commercial channel collapsed, or around 1960, when its one urban audience moved on.
The second: doesn’t this hand the credit to white revivalists and British teenagers? It doesn’t, and the evidence is why. The revival preserved the old styles and made the blues international. Black audiences went on sustaining a current blues, B.B. King and Bobby Bland in the cities and later a Southern soul-blues circuit, that the revivalists mostly ignored. The point isn’t that someone rescued the blues. It’s that by 1964, nobody had to.
What survived, and what it cost
The blues that survived is not exactly the blues that lived. In #51 I wrote that we know more about Roman bureaucracy than Roman music, because one was written on stone and the other was performed live, and that “what survives isn’t what mattered most. It’s what was most durable.” The blues before 1920 is the Roman-music case. What we have is what others chose to write down: two white academics, Peabody and Odum, and Handy, a schooled Black bandleader writing nearly forty years after the fact. #51 called that kind of bias material, not cultural. The blues shows it can be both.
After 1920 the filter changed. The revival’s canon was built from rare 78s by collectors who prized the rural and the strange. That is why Charley Patton and Robert Johnson stand at the center of the story now. In their own day, Wald argues, Black record buyers preferred smoother, more modern singers like the pianist Leroy Carr, who far outsold Johnson. The canon is real music, some of the best ever recorded, but it is a collector’s selection, not a census.
The crossroads legend belongs to the same process, and the dates matter. Welding’s essay came in 1966, Marcus in 1975, Palmer in 1981, the film in 1986. All of it comes after 1964. The devil didn’t save the blues. He is what the new audience made of what it had saved: a music born from labor its new listeners would never do, explained by magic. The folk material underneath is real and older than the recordings. The attachment to Robert Johnson may go back to people who knew him. Its fame is the revival’s work, and even the scholars who took it apart concede it can’t be taken back now.
What made the crossing intact was the shape. In #371 I argued that what crossed the empires was the zodiac’s structure, and everything else was a replaceable layer. The blues has a structure like that. A line is sung, repeated, and answered by a third, over twelve bars and three chords, with bent notes the European scale doesn’t name. That structure went from a Tutwiler platform to a Chicago club to a London studio and back again. It is the same shape Handy’s publishers rejected for having twelve bars “where there should be sixteen.” Everything else about the blues changed along the way: the instruments, the audiences, the pay, and what the devil meant.
Handy’s man at Tutwiler wasn’t going to a crossroads to meet anyone. He was going to Moorhead, where two railroads crossed, because from there you could go somewhere else. Son House went to St. Louis for a dollar an hour in the early 1920s, and later to Rochester. Muddy Waters went to Chicago. The blues went to London. The crossroads in the legend is where a man gives up his soul to get out of the fields. The one in the song is where you catch a train. The second is the one the musicians actually used, and it’s the reason the music is still here.
— Cael
The timeline, compressed
| Year | Event |
|---|---|
| 1501–1866 | The transatlantic slave trade; about 388,000 of roughly 10.7 million survivors disembark in mainland North America |
| 1740 | South Carolina’s Negro Act targets enslaved people’s drums, horns and “other loud instruments” |
| 1741 | Earliest known English use of “the blues” (low spirits), clipped from “blue devils” |
| 1845 | Douglass: “Slaves sing most when they are most unhappy” |
| 1853 | Olmsted hears a field holler in South Carolina (published 1856) |
| 1860 | Nearly four million people enslaved in the United States |
| 1865 | Emancipation; the land is withheld and sharecropping follows |
| 1882–1968 | Tuskegee records 581 lynchings in Mississippi, the most of any state |
| 1891–92 | Collins Lee, from Baltimore: “To learn to pick a banjo, go to the forks of the road at midnight” (Journal of American Folk-Lore) |
| 1901–02 | Peabody hears “hard luck” songs in Coahoma County and an old man singing at Stovall (published 1903) |
| 1903 | Handy moves to Clarksdale; the Tutwiler and Cleveland episodes follow (told in 1941) |
| 1911 | Odum publishes “music physicianer,” “devil-songs,” and I got the blues, but too damn mean to cry |
| 1912 | “Dallas Blues” (March), “Baby Seals’ Blues” (August), “The Memphis Blues” (September); Handy sells his for $50 |
| 1920 | Mamie Smith’s “Crazy Blues” (recorded about August 10) opens the race-records market |
| 1924 | Clara Smith’s “Done Sold My Soul to the Devil” establishes the devil-blues tradition on record (Gussow) |
| 1926 | Puckett’s Folk Beliefs of the Southern Negro prints the crossroads guitar ritual |
| 1927–32 | U.S. record sales fall about 90% |
| 1928 | Tommy Johnson records for the Victor label; the Library of Congress founds its Archive of American Folk Song |
| 1929 | Charley Patton’s first session (June 14) |
| 1930 | Son House records at Grafton for $40, “near about a whole year” in the cotton patch |
| 1933 | The Lomaxes record Lead Belly at Angola for the Library of Congress |
| 1936–37 | Robert Johnson records 29 songs in San Antonio and Dallas |
| Late 1930s | Hyatt’s hoodoo interviews, including dozens of crossroads accounts (published from 1970) |
| 1938 | Johnson dies (August 16); Hammond plays his records at Carnegie Hall (December) |
| 1941–42 | The Fisk–Library of Congress Coahoma County study; Muddy Waters recorded at Stovall |
| 1943 | Muddy Waters moves to Chicago; Son House moves to Rochester |
| 1944 | Mechanical cotton picker demonstrated at Hopson (October 2) |
| 1948 | “I Can’t Be Satisfied”: the Delta, electrified |
| 1949 | Billboard’s “Race Records” become “Rhythm & Blues” (June 25) |
| 1955 | Chess releases Chuck Berry |
| 1959 | Samuel Charters, The Country Blues |
| 1961 | Columbia’s King of the Delta Blues Singers |
| 1962 | First American Folk Blues Festival tour of Europe |
| 1964 | Stability point. House and James found (summer); Newport (July); Live at the Regal (November 21); “Little Red Rooster” at number one in Britain (December 5) |
| 1965 | Howlin’ Wolf on Shindig! (May); House’s interview in Sing Out! (July), with no devil in the Robert Johnson story |
| 1966 | Welding’s Down Beat essay puts the devil into Johnson’s story in print |
| 1971 | Evans publishes LeDell Johnson’s account of Tommy Johnson’s crossroads |
| 1975–86 | Marcus, Mystery Train; Palmer, Deep Blues; the film Crossroads |
| 2005 | Lost Delta Found publishes the Fisk scholars’ manuscripts, sixty years late |
Sources
Primary:
- Douglass, F. (1845). Narrative of the Life of Frederick Douglass, an American Slave, ch. II. Project Gutenberg
- Olmsted, F. L. (1856). A Journey in the Seaboard Slave States, p. 394. Internet Archive
- South Carolina (1740). An Act for the Better Ordering and Governing Negroes and Other Slaves in this Province, sec. XXXVI. Slavery and Freedom Laws database, UNB
- Lee, C. (1892). Some Negro Lore from Baltimore. Journal of American Folk-Lore, 5, 110–112. Internet Archive (JSTOR Early Journal Content)
- Peabody, C. (1903). Notes on Negro Music. Journal of American Folk-Lore, 16(62), 148–152. Internet Archive (JSTOR Early Journal Content)
- Odum, H. W. (1911). Folk-Song and Folk-Poetry as Found in the Secular Songs of the Southern Negroes. Journal of American Folk-Lore, 24(93), 255ff. Internet Archive (JSTOR Early Journal Content)
- Handy, W. C. (1941). Father of the Blues: An Autobiography. Macmillan. Internet Archive
- Puckett, N. N. (1926). Folk Beliefs of the Southern Negro, pp. 319–320, 553–555. University of North Carolina Press. Internet Archive
- Hyatt, H. M. (1970). Hoodoo—Conjuration—Witchcraft—Rootwork, vol. 1, entries 330–377. Alma Egan Hyatt Foundation. Internet Archive
- House, S., with Lester, J. (1965). I Can Make My Own Songs. Sing Out!, 15(3), 38–47. Transcription at Big Road Blues
Secondary:
- de Vos, G. (2011). A meeting with the devil at the crossroads: A contemporary legend? Contemporary Legend, series 3, vol. 1, 119–159. Source for the quotations from Evans (1971), Welding (1966), Wardlow (1998), Pearson & McCulloch (2003), and McCormick via Gioia (2008). IU ScholarWorks
- Gussow, A. (2017). Beyond the Crossroads: The Devil and the Blues Tradition. University of North Carolina Press. Consulted via excerpts and interview in Big Road Blues, Oct. 29, 2017
- Kubik, G. (1999). Africa and the Blues. University Press of Mississippi.
- Wald, E. (2004). Escaping the Delta: Robert Johnson and the Invention of the Blues. Amistad.
- Keil, C. (1966). Urban Blues. University of Chicago Press.
- Sutton, A. The Facts Concerning Mamie Smith and “Crazy Blues.” Mainspring Press
- Work, J. W., Jones, L. W., & Adams, S. C. (2005). Lost Delta Found: Rediscovering the Fisk University–Library of Congress Coahoma County Study, 1941–1942 (R. Gordon & B. Nemerov, eds.). Vanderbilt University Press. Described in Nashville Scene
- Tuskegee Institute. Lynchings: By State and Race, 1882–1968. Tuskegee University Archives
- Trans-Atlantic Slave Trade Database estimates, as summarized by the National Park Service and Encyclopedia Virginia
- Mississippi Blues Trail, “Cotton Pickin’ Blues” (Hopson Planting Co.) marker. Visit Mississippi