On the Trail of the Real John Henry


Mr. Garst is Emeritus Professor of Chemistry, University of Georgia, Athens and the author of "Chasing John Henry in Alabama and Mississippi: A Personal Memoir of Work in Progress" Tributaries: Journal of the Alabama Folklife Association (2002) 5: 92–129.

According to an old newspaper account (it was in the Atlanta Constitution) granite cutter Bill Hendricks appeared in court on September 1, 1913, to face charges of disturbing the neighborhood.  He had drunkenly "shouted and sung bad songs."  Bill admitted to one song, an inoffensive stanza of "John Henry" he had known since childhood.  He recited that stanza and said you sing it again if you want a longer song.  Convicted and fined $21.50, he "put the court on notice that it was a piece of malice on the part of the neighbors and not their objection to 'John Henry' that caused his arrest."

Bill was right.  "John Henry" is not a bad song.  To folklorists it is a "ballad," a story told in song.  Collections and recordings number in the many hundreds.  Aaron Copeland used the melody in his composition, "John Henry," for symphony orchestra.

John Henry is a hero to everyone, especially African Americans and labor-union members.  He is celebrated in novels, poems, cartoons, comics, paintings, sculptures, movies, etc., and he is reinvented.  He obviously inspired the creation of Steel, aka John Henry Irons, one of Superman's comic-book successors.

Why is so much made of a poor black laborer?  Because his mighty effort represents the best of the human spirit.

John Henry was a steel driver, a man who used a sledge hammer to pound steel drills to make holes in rock.  In tunnel boring the rock was blasted away by explosives packed into drilled holes.  John Henry said, "Before I'll let that steam drill beat me down / I'll die with my hammer in my hand."  With human skill, muscle, endurance, and determination, he drilled faster than a new-fangled, steam-powered machine.  "John Henry made fourteen feet / While the steam drill only made nine."  Then he died.  "He drove so hard that he broke his heart / He laid down his hammer and he died."

Is any of this true?  If so, or if the legend sprang from some other event involving a real person, who and where was he?

By 1933 Guy Johnson, Louis Chappell, and a few others had obtained over sixty versions of "John Henry" and a great deal of personal testimony.  Nothing in this mass of data can be assumed to be reliable.  Testimony and ballad versions vary wildly and are rife with contradictions. He died in at least ten states and Jamaica!  Received wisdom, from Johnson and Chappell, says he passed away in West Virginia.  MacEdward Leach writing some thirty years later said Jamaica. In this decade Scott Nelson has asserted death occurred in Virginia. I say Alabama.

In my opinion, West Virginia and Jamaica are no longer serious contenders.  Virginia and Alabama must duke it out over the historicity of "America's greatest single piece of folklore" (John A. and Alan Lomax).

In the Virginia story, the legendary steel driver was John William Henry, a convict at the old Virginia Penitentiary, Richmond.  He was leased in 1868 to work on the construction of the C & O railroad, 1868-72, at Lewis Tunnel, where he raced a steam drill and died.  His body was sent back to the penitentiary and he was buried in a mass grave by a white workhouse near a railroad.  A stanza of the ballad says that John Henry is taken to "the white house" and buried "in the sand" near a railroad.

There are just a few problems with the John W. Henry story.

First, "John Henry Something" was a very common name, much more common than "John Something Henry."  Accordingly, the historic person is more likely to have been "John Henry Something" than "John W. Henry."

Second, John W. Henry was 5' 1-1/4" tall.  How many little guys were famous steel drivers?  Probably none.

Third, because "John Henry" is so common, finding it among the more than two hundred convicts at Lewis Tunnel in 1870 is not unexpected and carries little logical force.

In the ballad, having John Henry buried at the "white house" is too good to be true.  Who could resist the idea that he was so important that he was put where our President, when he faced a difficult problem, could stroll out to visit his grave for inspiration?  If "white house" had been part of the original ballad, versions in which John Henry is buried elsewhere would not exist.  In fact, there are many.

There is no evidence at all that John W. Henry was a steel driver.  There is no evidence that anyone at Lewis Tunnel raced a steam drill.  Lewis Tunnel had its own graveyard.  Convicts who died there were probably buried there, not at the penitentiary.  Finally, no legend, testimony, or ballad version explicitly places John Henry at Lewis Tunnel.

If Virginia is not the right place, Alabama is.

In the Alabama story, John Henry was John Henry Dabney, an ex-slave from Copiah County, Mississippi, who followed Captain Frederick Yeamans Dabney to Dunnavant, Alabama, to bore tunnels through Oak and Coosa Mountains during the extension of the C & W (Columbus & Western) Railway from Goodwater to Birmingham, in 1887-88.  Dunnavant is just south of Leeds and about fifteen miles east of Birmingham and Red Mountain.  As Chief Engineer of the C & W, Captain Dabney was in charge of construction.  His father was John Henry's former owner.  When a steam-drill salesman came calling one day, Captain Dabney bet him that John Henry could beat his machine.  On September 20, 1887, he raced the steam drill outside the east portal of Oak Tunnel.  He won, fell out, revived, said he was blind and dying, called for his wife, and died by bleeding out from ventricular rupture as she cradled his head.  As the ballad says, "He drove so hard that he broke his heart."

The Alabama story derives mostly from testimony.  Unlike most John Henry testimony, that which places him in Alabama is, within reason, coherent.  Three informants said that he worked for a Dabney, two that both he and Dabney were from Mississippi, three that he worked at Coosa and Oak Mountains, and two that Red Mountain was nearby.  C. C. Spencer, who claimed to have been an eyewitness to John Henry's race and death, gave many details, some of which have been verified.

Spencer's description of John Henry's death led Steven Harris, M.D., to offer ventricular rupture as a likely cause of death: "No, strokes don't do this. You can get blindness with a posterior vertebral stroke, but it shouldn't kill you right off -- or make you unconscious.  And unconsciousness which reverses when the person is laid down is classic for blood loss shock.  As is blindness and a roaring in the ears (all low blood pressure things) .... And chest pain would precede, from the ischemia of the heart attack itself ...."  Versions of the ballad mention that John Henry had chest pain, blindness, and roaring in the ears.

A Dabney family memoir mentions a slave boy named Henry who was a teenager during the Civil War.  Consistent with this, Henry Dabney, age 20, married, "works on farm," is enumerated in the 1870 federal census for Copiah County.

A strong legend around Dunnavant makes the east portal of Oak Tunnel the site of John Henry's contest.  For years a steel drill stuck up from the rock there.  It was said to have been John Henry's last "steel," a memorial to him.

This comes close to nailing John Henry down in Alabama, but there are some loose ends.  We may have to drive a stake through his heart to stop his perambulating and get rid of his ghost at Big Bend Tunnel and those other places.

That might be possible.  A candidate for John Henry's grave lies just outside the fence (whites only inside, I suppose) enclosing Sand Ridge Cemetery ("buried him in the sand"), Dunnavant.  This cemetery is within sight of the C & W tracks about a mile across a valley and is on a "white road" (mentioned in one version of the ballad).

According to some versions of "John Henry," his hammer was buried with him.  Ground-penetrating radar might reveal it (nine-, twelve-, twenty-pound?)  Exhumation might permit medical and genetic examinations as well as provide an opportunity for staking.  Perhaps he has living relatives in Mississippi with whom DNA samples could be compared.

The Georgia Historical Society owns massive records of the Central of Georgia system, to which the C & W belonged.  They are uncataloged, in need of conservation, and inaccessible.  They might contain C & W records that mention John Henry, steam drills, or the contest.

I invite historians, the more the merrier, to join the search for John Henry in Alabama.  The application of your talents could lead to paydirt.

(For more Alabama information, see Tributaries 2002 and online "John Henry" discussions at Mudcat Cafe, Ballad-L, and Pre-War Blues, Yahoo group.)

Response by Scott Nelson (William & Mary)

John Garst would like to believe that John Henry is from his town in Alabama, as would the Alabama Folklife Association, but there is simply no coherent evidence to prove it. What he largely does is poke holes in my own work – Steel Drivin’ Man, published last month by Oxford University Press – rather than doing the archival work that historians do to prove their point. His latest attack on me was posted on History News Network. He has also attacked me and my work on websites, on amazon.com, and on numerous blogs.

Let me rebut his criticisms of my work point by point.

Name: “John Henry” was actually an uncommon name among black men in the South in the 1870s. There were fewer than a score of black John Henry’s in the 1870 census, most too young or too old to be candidates. John W. Henry of Virginia who I discuss in my book Steel Drivin’ Man is listed both in penitentiary records and in the Virginia census as John Henry. Professor Garst would like to believe that John Henry is “John Henry Something Else”, but there is no direct evidence that suggests this.

Height: In the nineteenth and early twentieth century, tunnel workers – particularly the men who drilled holes for tunnel blasts and in hard rock mines – were very small men. This was a necessity because the holes produced by nitroglycerin rock blasts were quite small. These holes were only widened to man-sized holes after the initial drilling was completed. We should not be surprised to find that John Henry was short in stature; we should expect it if he were a tunnel driller.

Commonness: As I said above, John Henry was an uncommon name for a black man in the 1870s. By the 1880s, after the events that I describe in my book, it had become a much more common name.

White House: Professor Garst suggests that it was irresistible to put John Henry at the President’s “White House” and that my explanation for this version of the song – that the white house refers to a building at the Virginia Penitentiary – is “too good to be true.” In fact, the discussion of the white house in the song is peculiar. Both Johnson and Chappell noted it and found it confusing. In fact, the use of the term “White House” to describe the President’s residence became common only after Theodore Roosevelt became president in 1901. If the song was composed before 1901, and it certainly was, then we must explain why many versions of the song end with a discussion of the white house. My book explains this.

Lewis Tunnel: Professor Garst plays fast and loose here, arguing that no one “raced” a steam drill at Lewis Tunnel and that Henry was not a steel driller. I have presented considerable evidence that two different steam drills were used at this tunnel in the 1870s. My book demonstrates how and when John Henry came to Lewis Tunnel. I demonstrate why, contractually, all convicts had to be returned to the penitentiary dead or alive so that the contractors avoided a one-hundred dollar fine. I show how the penitentiary surgeon attributed many of the deaths of the convicts to work on the Lewis Tunnel. I quote from two different contemporary sources that describe workers and steam drills working side-by-side. I do not have a photograph of John Henry fighting the steam drill, but I am sure that even then Professor Garst would dispute it.

Burial: Yes, Lewis Tunnel has its own graveyard. The nineteenth century headstones refer only to white men and women in it. It is improbable that black convicts would have been buried in a white cemetery near Lewis Tunnel. To avoid the fine discussed above, all prisoners had to be returned dead or alive to the penitentiary. This is attested to by the surgeon’s report at the Virginia Penitentiary.

The remainder of Professor Garst’s work, like his article in Tributaries, gives no evidence of steam drills used near his home in Alabama, or even a description of the tunneling work that took place. He has no evidence that John Henry Dabney worked for any contractor on this road. He has oral testimony that was provided by and then dismissed by Chappell and Johnson more than 60 years ago. Chappell and Johnson received only two witnesses that placed the conflict in Alabama and dozens and dozens that placed the conflict on the Chesapeake and Ohio railroad. Beyond these quotations reproduced from the two books, Garst provides a link to the census that proves that a black man named Henry Dabney is listed as working on a farm near the site. The rest is supposition. Professor Garst has done no archival research to make his case, and there are numerous materials available including company reports of the C&W, engineering reports, and local newspapers available to him.

It is wonderful that scholars of all stripes are interested in the legend of John Henry, and would never knock down someone who has an interest in the topic. But as Professor Garst knows, we can not simply will a chemical reaction to take place, and hope alone does not make Alabama the place where John Henry died.

comments powered by Disqus

More Comments:

John F Garst - 10/3/2007

A PDF file containing the text of my talk in Leeds, AL, on September 15, 2007, can be downloaded from the WWW site of the Alabama Folklife Association.


On the left side, toward the top, is a link entitled "John Henry in Leeds: Celebrating a Local Legend." Click on this to bring up a page devoted to the September 15 event.

On the right side of that page, perhaps a third of the way down, is a picture of me speaking. The last thing in the accompanying text is
"For text of John Garst talk, click here." Clicking there will results in a PDF download.

John F Garst - 9/20/2007

In my talk in Leeds, Alabama, on September 15, 2007, I claimed that it is "beyond reasonable doubt" that John Henry was a real person who was at Dunnavant, Alabama, in 1887. "Beyond reasonable doubt" is the standard applied in our criminal courts. If the case could be brought to trial, I believe that Dunnavant could be convicted of having had John Henry there.

I would have posted the text of my talk here had it not been for HNN's copyright policy. I will try to post it elsewhere. If I succeed, I will post a notice here.

I note that Dunnavant is the only proposed location for John Henry that has a long and strong local legend that has confirming ties with coherent testimony, documentation, and ballad texts.

Big Bend Tunnel, West Virginia, has a local tradition but only incoherent testimony, no documentation, and only one confirming tie to ballad texts (the explicit mention of Big Bend Tunnel).

Lewis Tunnel, Virginia, has some weakly suggestive documentation but no tradition, no testimony, and no confirming ties to ballad texts.

John Garst

Don Clowers - 8/25/2007

If you can not prove that the ledendary J. Henry actually existed, how can you hope to prove or even suppose that he was or has been anywhere in the vicinity of Leeds. AL or anywhere else? Is that not like the old axiom “putting the cart before the horse”? That Dr. Garth was my point in my original contention or as the Leeds News refers to it as “criticism”.

John F Garst - 3/28/2007

I wrote:

"Ballad: John Henry died on a Tuesday.
"Utah: John Henry died on September 20, 1882.
"Fact: 1887 was the only year the C & W was under construction in September.
"Fact: September 20, 1887, was a Tuesday.

"The probability that this agreement is coincidence is 1/7, 14%. Therefore the probability that it is not coincidence is 86%, which can also be taken as the probability that, in fact, John Henry died on September 20, 1887, at Dunnavant, Alabama."

Although the probability (based on this evidence alone) that John Henry died on September 20, 1887, may be in the vicinity of 86%, this line of reasoning is erroneous. That the probability of agreement between date and day is 1/7 if either or both are random does not properly imply that an observed agreement has a 6/7 probability of being nonrandom.

Other treatments of the problem, however, do indicate a high probability that the date and day are not random. The most obvious way for them to be nonrandom is that they reflect historic truth.

John F Garst - 1/3/2007

For the record, I wrote:

"Indeed, the proper understanding of frequency in a mature, history-based ballad tradition may be counterintuitive. It is summarized in The Principle of Frequent Crap and Rare Truth: ballad elements that are found frequently are often crap, that is, not in the original ballad, while elements found rarely are often original. To the extent that the original ballad reflects historic events this implies that rare elements are more likely to be historic than frequent ones."

At the Google Group rec.music.country.old-time, Joseph Scott challenged "rare elements are more likely to be historic than frequent ones." After some discussion, I decided that I could not logically defend either this statement or its opposite. I still think it likely to be true but because I can't defend it well I have retracted it.

As I wrote at rec.music.country.old-time,

"The working principle is

(1) Don't assume that a rare element was absent from the original ballad.

(2) Don't assume that a common element was present in the original ballad."

John F Garst - 12/30/2006

The ballad itself contains phrases and lines that support the Alabama story. Before listing some of these, it should be noted that each has been recovered rarely, most only once. If Nelson were correct in believing that frequency of recovery is evidence in support of the recovered motif, these would need to be ignored. However, as noted above, frequency is not good evidence.

Indeed, the proper understanding of frequency in a mature, history-based ballad tradition may be counterintuitive. It is summarized in The Principle of Frequent Crap and Rare Truth: ballad elements that are found frequently are often crap, that is, not in the original ballad, while elements found rarely are often original. To the extent that the original ballad reflects historic events this implies that rare elements are more likely to be historic than frequent ones.

As ballads are passed on in oral tradition they mutate. Many factors contribute to mutation, including mishearing and memory failures. There is also a controlling artistic consideration: each singer wants to sing a good story. Each singer in the chain of transmission will retain and likely pass on those elements of a ballad that he or she finds particularly attractive. The truth of what is sung will not be a consideration. Indeed, once a ballad is a few steps removed from its source most singers will not know anything about the historic truth behind it. Singers will drop dull factual material, change it to something more attractive, or introduce attractive elements from other sources.

Thus, by losing non-attractive elements and gaining attractive ones, a ballad progresses toward one or more stable end points in which only attractive elements remain. Thus, elements found frequently will often not have been present in the original ballad. “White house” may be such an element.

In “John Henry,” “Big Bend Tunnel on the C & O road” is frequent and is therefore suspect. Similarly, “Polly Ann” (as John Henry’s wife/woman) is frequent. I’d be willing to bet that it is not historic.

Rare elements of a ballad are non-attractive. Otherwise they wouldn’t be rare. It is likely, therefore, that a rare element is a remnant of the original. For history-based ballads, this implies that rare elements often will be historically correct. When a rare element coincides with testimony or fact, its probability of being historically correct is greatly enhanced.

Below are some rare but relevant lines from recovered versions of “John Henry” and agreeing testimony or fact.

Ballad: V’ginny gave him birth (refers to “John Henry’s cap’n Tommy”).
Fact: Captain Dabney was born in Virginia. He moved to Mississippi by age one.
Plausibility: “Tommy” is a transmissional mutation of “Dabney.”

Ballad: ’tween them mountains/between two mountains (refers to John Henry).
Fact: The Dunnavant Valley, Alabama, lies between Coosa and Oak Mountains.

Ballad: John Henry died on a Tuesday.
Utah: John Henry died on September 20, 1882.
Fact: 1887 was the only year the C & W was under construction in September.
Fact: September 20, 1887, was a Tuesday.

The probability that this agreement is coincidence is 1/7, 14%. Therefore the probability that it is not coincidence is 86%, which can also be taken as the probability that, in fact, John Henry died on September 20, 1887, at Dunnavant, Alabama.

Ballad: Maggadee was her name (refers to John Henry’s wife/woman).
Fact: Henry Dabney married Margaret Foston in December, 1869.
Plausibility: “Maggie D” was a nickname for Margaret Dabney.
Plausibility: “Maggadee” originated as “Maggie D.”

Regarding “Polly Ann,” I offer the following plausible chain of mutations from “Margaret Dabney” and “Maggie D.” As the ballad is propagated “Maggie D” is heard as “Maggadee,” which is found in tradition (one version): “John Henry had a little woman / Maggadee was her name / When John Henry took sick and had to go to bed / Maggadee drove steel like a man.” “Maggadee” sounds like and suggests “Magdalene,” which would be incomplete without “Mary.” “Mary Magdalene” is found in tradition (a couple of versions). “Mary Magdalene” is somewhat awkward and would trend naturally to “Mary Ann.” In addition, “Mary Ann” provides a good rhyme: “John Henry had a little woman / Her name was Mary Ann / When John Henry took sick and had to go to bed / Mary Ann drove steel like a man.” “Mary Ann” is found in tradition (a few versions). “Polly” is a nickname for “Mary.” Thus, “Polly Ann” arises. “Polly” is very common in balladry and represents an attractive feature. Thus, “Polly Ann” becomes the stable end point of the feature of the ballad naming John Henry’s wife/woman. It is found very frequently.

The John Henry tradition placing him at Dunnavant, Alabama, first appeared in print in 1930, 43 years after the alleged event. However, it is said to have existed among Central of Georgia railroad employees and citizens of the Dunnavant and Leeds well before that. This tradition could date from the historical event itself.

John F Garst - 12/20/2006

Scott Nelson states that I have not examined "company reports of the C & W, engineering reports, and local newspapers available to" me. Actually, I have spent many hours with the usual bad microfilms of the possibly relevant newspapers from Alabama and Mississippi that I have been able to locate. My Tributaries article, "Chasing John Henry," cites six items from these newspapers. I have not seen many C & W papers because I haven't been able to find them. Nelson has graciously given me some tips on tracking down these records. I am grateful.

I had hoped that Nelson would turn his research attention to Alabama railroads and the question of John Henry there (the "Alabama story" about John Henry Dabney). However, he tells me that he deems his work on John Henry complete (the "Virginia story" about John William Henry) and has already redirected his research. I intend to continue pursuing evidence pertinent to the Alabama story.

Before replying to Nelson's comments about the evidence for the Virginia and Alabama stories, I mention thoughts that have recently come to my attention from others.

First, I am not the only one to opine that the evidence for the Virginia story, presented in "Steel Drivin' Man," is weak. William Grimes (New York Times) asks, "Do the facts add up?" and replies, "Maybe." To Jon Sobel (blogcritics) the evidence is "inconclusive."

Second, "Aunt B.", at her "Tiny Cat Pants" WWW site, said that she considers the most compelling thing about the Alabama story to be that John Henry has a wife. In the Virginia story, John Henry seems to be single. “Aunt B.” thinks that John Henry’s wife’s “picking up the hammer and driving steel like a man is one central part of the song."

That’s true, and John Henry's wife/woman also plays other roles in the ballad. She appears frequently as the mother of his children, as a distraught wife (after he collapses), and as a widow.

If I thought the frequency of a motif in tradition determined its reliability as history, I'd consider John W. Henry’s single status to be strong evidence against the Virginia story. I don't think that, so I consider it to be less-than-strong evidence.

Some might not think it evidence at all, arguing that a ballad or legend may be largely tall tale. Ballads and legends do accrue bullshit and lose truth, but it is hard to imagine that a history-based ballad or legend will not retain some true elements. For example, according to “Frankie and Johnny,” Frankie shot “her man” Johnny or Albert three times in a barroom, hotel room, or brothel after he had “done her wrong” with Nelly Bly. In fact, Frankie Baker didn't shoot either Johnny or Albert but she did shoot "her man," Allen Britt. She didn’t shoot him three times after he had “done her wrong” with Nelly Bly but she did shoot him, once, at home, after he had been with Alice Pryor.

Because ballads and legends are unreliable, it is tempting to a truth-seeker to disregard them altogether, but because history-based ballads and legends are likely to contain truth, that would be a mistake. Legend should be winnowed to separate truth from bullshit. That may not be easy but it is not impossible.

Because the a priori probability of the truth of reports that John Henry had a wife is not zero, John W. Henry's lack of one is evidence that he is not the legendary steel driver. It may be weak evidence, but weak evidence should not be disregarded because many pieces can accumulate to a level of significance.

Third, Douglas Galbi, who studies history through the lens of psychology, doubts the Virginia story because John W. Henry was a convict laborer. "John Henry obviously had great respect for himself and for his work. That's the sort of attitude I could easily imagine in a freed black slave ... being in prison is hugely demoralizing. A convict in a work gang just wouldn't have the sense of self to do what John Henry did. Persons who come into prison have that sort of quality pressed out of them. There's no way, in my opinion, that John Henry could have been a convict laborer."

The legendary John Henry is indeed proud. He said to the Captain, "A man ain't nothing but a man / But before I'll let that steam drill beat me down / I'll die with my hammer in my hand." In the Virginia story such an exchange would be very unlikely.

John Henry's pride in his ability is the psychological core of the ballad and legend. How reasonable is it to imagine that the historic John Henry lacked this pride?

The Virginia story is inconsistent with many elements of the John Henry tradition. Thus, the prison setting would hardly be one in which the Captain loved “John Henry like his only son," as is stated in one ballad version. In the Alabama story this is a distinct possibility. It is likely that Captain Dabney, who was about 15 years older, knew John Henry from birth.

Nelson's comments begin with the statement that I "would like to believe that John Henry is from" my "town in Alabama, as would the Alabama Folklife Association." Evidence is what counts, of course, not “likes.”

Toward the end of his response Nelson refers again to my "home in Alabama." For the record, I live in Athens, Georgia. I was born and raised in Jackson, Mississippi, and I have never lived in Alabama.

Regarding the Alabama story, Nelson states, “… there is simply no coherent evidence to prove it." In fact, coherence is its strength. The testimony supporting the Alabama story is reasonably self-consistent and consistent with documentation. Neither the West Virginia (Big Bend Tunnel) nor the Virginia story is as coherent.

The evidence for the West Virginia story is full of inconsistencies and contradictions (see “Chasing John Henry” and Johnson’s and Chappell’s books). Johnson admitted that there was a “mountain of negative evidence” against his conclusion favoring Big Bend Tunnel. Chappell’s summary is mostly speculation. Neither investigator adequately addressed the Alabama testimony.

Neither even heard about the Virginia story. Considering that they interviewed men who had worked at Lewis Tunnel, that failure itself suggests that the Virginia story is wrong.

As for proof, mentioned by Nelson, no one will ever have it in the deductive, mathematical sense of that word. We all seek the story that best agrees with what is known.

Based on census reports for 1870 and 1880, Nelson challenges my assertion that "John Henry Something" was more common than "John Something Henry." I could be wrong but it doesn’t much matter. Even if I’m right it is just one piece of weak statistical evidence.

Even so, I don’t think much of the census as a source of information of this type. Bob Eagle, an Australian lawyer and blues researcher who uses the census extensively and who has about eleven years of experience doing so, sent an e-mail from which I quote (in part), with his permission. "To take the presence or absence of 'John Henry' in a particular census as any indication of reality is to be hopelessly naive. There are few 'John Henry' listings coupled with *any* surname in 1870, but many, many more in 1880 (and not just children, either). This could just represent a change in policy in Washington, DC."

I accept Nelson's point that very small men could have been valuable steel drivers in cramped quarters. It remains hard for me to imagine that a man just over five feet, one inch tall could have been a noted steel driver, a winner of competitions against larger men. John Henry is celebrated for his steel-driving prowess. To me John W. Henry's small size is one among many factors that make it unlikely that he inspired the ballad.

The case for the Virginia story rests on the historic truth of a line found occasionally in the ballad, "They took John Henry to the White House" (not always capitalized). Of 58 independent versions of the ballad published in the books of Guy Johnson and Louis Chappell, eight contain a line like this. They are from West Virginia (3), Kentucky (2), Virginia, Ohio, and Georgia. Eight other versions specify another place for John Henry's burial ("burying ground," "graveyard," "river," "father's house"). These are from West Virginia (4), Virginia, Tennessee, Alabama, and Louisiana. Both types are widespread. As far as these data permit judgment, it seems that West Virginians and Virginians were equally likely to sing of John Henry's burial at the "white house" and somewhere else.

Nelson notes that the President's house became widely known as the "White House" only after 1901. The relevance of this is not clear. We have no version of "John Henry" that is known to predate about 1912. "White house" could have entered the known ballad tradition after 1901.

Even if "white house" were in the ballad before 1901, I think singers would have found it attractive and tended to pass it on in oral transmission. It would have been unusual and it sings well. It makes more sense to me to imagine that burial in a "graveyard" or "burying ground" was in the original ballad, and that "white house" is a product of mutation, than to imagine the reverse.

Because there is no evidence that "white house" was in the original ballad, the identification of John W. Henry as the legendary steel driver is shaky at best. My opinion is that "white house" is unlikely to have been in the original. If it were not in the original, then the Virginia story would have no foundation.

If "white house" were not in the original ballad, how would it have gotten into later versions? Nelson could be entirely correct. To the first singers of "white house" the reference could have been to the white workhouse at the Virginia Penitentiary. "White house" was probably part of local jargon referring to that penitentiary. Indeed, I once came across a folk song in which it was clear that "white house" referred to a penitentiary or jail. Having John Henry buried there is easily understood as part of the relocalization of the ballad from Alabama to the West Virginia/Virginia area.

In an early version of "John Henry" from Florida, the stricken steel driver is taken "on that long white road." In Virginia and West Virginia, "white road" could have been a seed that was replaced by "white house," thus inserting a reference to the Virginia Penitentiary.

In Dunnavant, Alabama, a white road (sand) leads to Sand Ridge Cemetery, just outside of which lies a grave that could be John Henry's. It is about a mile across a valley from the C & W tracks with a clear view between. If the candidate grave were John Henry's the facts would conform to ballad stanzas containing something like "...they buried him in the sand / An' every locomotive come rolling by / Says, 'There lies that steel-drivin' man,'" not all versions of which refer to the “white house.”

Not all "white house" versions of "John Henry" have him buried there. In a West Virginia version he is taken from the white house to the tunnel and later buried in the sand of the "new burying ground." In another, "People came from the White House" to watch John Henry drive steel. In a Kentucky version he "left the white house" to drive steel. In a version of unknown origin, published in 1915, John Henry is brought "from the white house" and taken "to the tunnel to drive." These instances illustrate the inconsistency of the role of the "white house" in the ballad.

In Nelson's response to my essay, he does not address the facts that there is no evidence that John W. Henry was a steel driver and no evidence of a contest between man and machine. The extent of what he has established is that manual and steam drilling occurred simultaneously at Lewis Tunnel for about a year. Steam drilling was abandoned eventually and the job finished with manual drilling.

A Lewis Tunnel worker who operated the engine supplying steam to the drills named Bob Jones as the best steel driver there. He did not mention John W. Henry.

As presented in "Steel Drivin' Man," the evidence that John W. Henry died at Lewis Tunnel is weak. He could have escaped. The book does not mention records of escaped convict workers.

I have tried hard to make what Nelson writes in "Steel Drivin' Man" imply that convicts' bodies were sent back to the Virginia Penitentiary for burial but I have been unable to complete the required logical chain. "Steel Drivin' Man" states (p 78), "Contractors like Mason would have to post a bond that guaranteed the safe return of prisoners ... Governor Wells wrote that contracts have 'stipulated damages of one hundred dollars for each prisoner not returned.'" I cannot understand the return of a corpse as a "safe return" and I have trouble construing "one hundred dollars for each prisoner not returned" as referring to the dead. Later (pp 89-90) it is stated, "The surgeon's report for the penitentiary listed the names of the men who died in prison but only gave the total number of men who died on the C&O Railroad." I see nothing here that implies that the bodies of those who died at Lewis Tunnel were returned. The surgeon could have been notified of deaths, and enumerated them, without receiving the bodies. On p 92 it is stated, "Nearly one hundred men came back between 1871 and 1873, most of them dead." There is no note or citation of a reference. The citation at the following sentence provides no quotation or description of content. It is not clear why it is thought that any corpses came back between 1871 and 1873. It would have been a waste of time, money, and effort to send them back to the penitentiary, especially when, according to an eye-witness, there was a graveyard at Lewis Tunnel. If there is something in the surgeon’s report that clarifies this, it does not appear to have been included in “Steel Drivin’ Man.”

I don't know whether or not the 1870s graveyard at Lewis Tunnel was the Lewis Tunnel Cemetery that exists today. If it is, Nelson's inability to find blacks there could be because their graves are unmarked. If whites are buried there, as he states, blacks would surely have been buried in a different section, probably in an area not considered to be part of the white cemetery.

Nelson comments that my article in Tributaries fails to give evidence for various claims. This raises the question, "What is evidence?" Inconsistently, in my opinion, Nelson accepts as evidence ballad lines of questionable historic truth but rejects the testimonies three credible witnesses (not two, as he stated) who place John Henry's contest in Alabama. Further, he appeals to the alleged fact that "dozens and dozens" of witnesses place "the conflict on the Chesapeake and Ohio railroad," as if frequency of recovery of testimonial information were a reliable guide to truth. It is not.

Johnson and Chappell made the mistake of placing John Henry at Big Bend Tunnel because they were impressed by the strength of that tradition. Despite the "dozens and dozens" who put John Henry at “Big Bend Tunnel on the C & O road,” "Steel Drivin' Man" has him at Lewis Tunnel. Thus, Nelson cites frequency of testimony as evidence favoring John Henry on the C & O but ignores similarly frequent testimony that he was at Big Bend Tunnel.

Frequency of recovery of elements of a traditional legend, whether in song or testimony, is at best suggestive. It is certainly not definitive. The American murder ballad “Ella Speed” is based on a New Orleans event. However, testimony from singers placed it in Dallas. Similarly, the murder ballad “Delia”/”Delia(‘s) Gone” has taken such hold in the Bahamas that it is regarded there as a national song based on a local event. In fact, Delia Green died in Savannah, Georgia, in 1900. There once was a “Delia” ballad tradition in the Savannah area but it seems to have disappeared long ago. As far as I know, this ballad is known today in Savannah only in forms based on those that re-entered the United States from the Bahamas in the 1950s and thereafter.

As to Nelson’s “dozens and dozens” who “placed the conflict on the Chesapeake and Ohio railroad,” the actual number of credible witnesses for the C & O is far fewer. In the 1920s Johnson and Chappell interviewed about a dozen men who had worked on the construction of Big Bend or Lewis Tunnel. Of these, about half thought the legendary John Henry had been at Big Bend Tunnel. Some of the others were adamant in their belief that John Henry had not been there – if he had been they would have known about it. Only a half-dozen or so of these men placed John Henry at Big Bend Tunnel, the only spot on the C & O where anyone placed him, and they are contradicted by an equal number of others.

The one man who said that he had seen part of John Henry’s contest with the steam drill at Big Bend Tunnel was a very poor witness. He evaded giving details by saying that it was not a big deal and that he had just looked in on it occasionally as he carried out his duties of carrying water and steel.

As for Lewis Tunnel, I know of no testimony or tradition that puts John Henry there. A few informants who thought John Henry had been at Big Bend Tunnel incorrectly named C. R. Mason as the construction contractor. In fact, the contractor at Big Bend Tunnel was W. R. Johnson. Mason was at Lewis Tunnel. This is the closest to a Lewis Tunnel John Henry tradition that I know of. If John Henry really died at Lewis Tunnel, why has there never been a tradition that says so?

According to Nelson, the oral testimony on which the Alabama story is based “was provided by and then dismissed by Chappell and Johnson more than 60 years ago.” Actually, it is written testimony, and to say that Johnson and Chappell “dismissed” it is misleading.

Johnson was so concerned about the Alabama testimony that he made a considerable effort to check it out. He turned eventually to Big Bend Tunnel because he couldn’t find the relevant place in Alabama (“Cruzee,” “Cursey,” or Oak Mountain) and he thought Big Bend Tunnel had the strongest tradition. As it turns out, he made two mistakes, not finding Alabama’s Coosa Mountain and assuming that Big Bend Tunnel had the strongest tradition. If he had found the Alabama location, the community of Dunnavant, he would have found there a tradition as old and strong as that for Big Bend Tunnel. Unfortunately, it was after Johnson’s book was published (1929) that the Alabama tradition first came to light in print (1930).

Chappell’s book was published in 1933 but it is obvious from its contents that he was unaware of the 1930 article. Referring to Alabama, he wrote, “The alleged drilling-contest in particular is still a heavy item … the reports of Henry’s connections in that section … will probably deserve more attention as authentic records when the place where the alleged contest occurred is found in Alabama.” That place was found in 2001, and I have been trying to give it the attention Chappell called for.

Chappell made an effort to identify “Cruzee” Mountain with the Santa Cruz Mountains of Jamaica. He found a John Henry tradition there, and a very useful one to me, but he was convinced that it was not the primary tradition. I agree.

Nelson stated that I give “no evidence of steam drills” used in the construction of the C & W extension. In “Steel Drivin’ Man” he dismisses Big Bend Tunnel because no steam drills were used there. This is not an adequate reason. Both Johnson and Chappell knew that steam drills were not used at Big Bend Tunnel but they persisted in favoring it as the site of John Henry’s contest because they imagined the possibility that a steam drill was brought there for evaluation, as more than one informant testified. Similar testimony says that the steam drill was being tested against John Henry in Alabama, so the question whether or not steam drills were actually used in construction, at either place, is irrelevant. As a matter of fact, however, after “Chasing John Henry” was published my attention was drawn to a printed account in which someone states that an ancestor provided wood for the steam drills used in building Oak and Coosa Tunnels.

Here is a recap of the evidence for the Virginia story.

A young, black, Virginia Penitentiary convict, John W. Henry, who was just over 5’ 1” tall, was probably leased by the State of Virginia to work on the building of Lewis Tunnel, where hand and steam drilling were used simultaneously for about a year, after which the steam drills were abandoned. Penitentiary records do not mention that he returned.

At the Virginia Penitentiary there was a white workhouse alongside which was a mass grave. A railroad ran nearby. The soil in which the grave was dug was not sandy. However, burial boxes in the grave were separated by thin layers of sand.

A stanza of the ballad, “John Henry,” found occasionally, runs as follows.
They took John Henry to the White House,
And buried him in the san’,
And every locomotive come roarin’ by,
Says there lays that steel drivin’ man.

The Virginia story accounts for this stanza by making questionable assumptions, that John W. Henry was a steel driver (no evidence), that he was a noted steel driver (no evidence and unlikely), that he raced a steam drill (no evidence), that he died at the tunnel (not the only possibility), that his body was returned and buried at the Virginia Penitentiary (no clear evidence that this was the practice), that a ballad maker would describe him as buried “in the sand” when the grave was in soil that was not sandy (doubtful), that John W. Henry would have been singled out among steel drivers at Lewis Tunnel for immortalization (very doubtful), and that a Lewis Tunnel tradition might never materialize, despite its being the place where it happened (very doubtful). In addition, if the “white house” reference in the ballad is an artifact, which I consider very likely, then the whole scheme of evidence collapses. Further, the Virginia story is inconsistent with the psychological core, and runs counter to some of the central themes, of the ballad and legend.

The evidence for the Alabama story is too extensive to review in detail. Here are especially relevant elements of the testimonies (paraphrased) of five informants. These testimonials come from various parts of the United States and Jamaica and were given at least 40 years after the events to which they refer. The informants lived in Utah, Birmingham (Alabama), Michigan, Jamaica, and Leeds (Alabama). Four gave their testimonies to Johnson and Chappell, probably in 1927-32. The testimony of the Leeds informant was published in 1955.

Utah: I was there. I was about 14 years old.
Birmingham: I knew John Henry.
Birmingham: John Henry was the champion of the world with a hammer.
Michigan: My uncle worked with John Henry.

Utah: It was near Red Mountain.
Birmingham: I was driving steel on Red Mountain.

Utah: It was 1882.
Birmingham: It was somewhere about 45 years ago. (about 1882)
Michigan: It was 1887.

Utah: It was on the Alabama Great Southern railroad.

Utah: John Henry was at Cruzee Mountain.
Birmingham: John Henry was at Cursey Mountain.
Michigan: John Henry died at Oak Mountain.

Utah: John Henry worked for Dabner.
Michigan: John Henry worked for Dabney.
Jamaica: John Henry worked for Dabner.

Utah: John Henry was from Holly Springs, Mississippi.
Utah: He was John Henry Dabner.
Utah: John Henry was probably a slave in the Dabner family in Mississippi.
Leeds: Both John Henry and his boss were from Mississippi.

Utah: I saw John Henry collapse and die.
Utah: He said, “Send for my wife, I am blind and dying.”
Utah: His wife came and cradled his head.
Birmingham: John Henry beat the steam drill while I worked at Red Mountain.
Michigan: My uncle saw John Henry beat the steam drill and die.

Here are some documented facts.

Red Mountain is at Birmingham, Alabama.
The Columbus & Western extension was built in 1887-88.
The C & W was owned by the Central Rail Road and Banking Company of Georgia.
The latter later became the Central of Georgia.
The C & W extension went through Dunnavant and Leeds, Alabama.
Dunnavant and Leeds are about fifteen miles east of Red Mountain.
The C & W crossed Coosa and Oak Mountains at Dunnavant.
Coosa and Oak Mountains are parallel ridges two miles apart.
The C & W bored tunnels through Coosa and Oak Mountains.
The Chief Engineer for the C & W was Captain Frederick Yeamans Dabney.
Captain Dabney was in charge of the construction of the C & W extension.
Captain Dabney was born in Virginia but moved to Mississippi by age one.
Captain Dabney’s father lived in Raymond, Hinds County, Mississippi.
Captain Dabney’s uncle owned Burleigh Plantation, Hinds County, Mississippi.
Burleigh was between Raymond and Crystal Springs, Copiah County.
Captain Dabney grew up in Raymond but visited Burleigh frequently
In 1860 Captain Dabney’s father, a lawyer, owned eight slaves.
In 1860 Captain Dabney’s uncle owned 154 slaves.
After the Civil War, Captain Dabney lived in Crystal Springs.
Captain Dabney’s father owned a slave boy named Henry.
During the Civil War, this Henry was a teenager.
Henry Dabney, age 20, appears in the 1870 Copiah County census.
Henry Dabney is listed as working on a farm.
Henry Dabney married Margaret Foston in December, 1869.

Comparisons of testimony with documented facts reveal remarkable agreement. The disagreements (Holly Springs vs Crystal Springs; Alabama Great Southern vs Columbus & Western; 1882, somewhere about 1882, 1887 vs 1887-88; Dabner vs Dabney; Cruzee, Cursey vs Coosa) can be plausibly attributed to memory lapses (Mississippi “Springs” town, railroad line, date) or misreproductions (Dabner, Cruzee, Cursey).

The Henry Dabney of the 1870 census is the right age to have been Captain Dabney’s father’s slave boy. His age and marital status are appropriate for the John Henry Dabney of Alabama in 1887. Alternatively, John Henry Dabney could have been one of Captain Dabney’s uncle’s many slaves.

The John Henry tradition placing him at Dunnavant, Alabama, existed among Central of Georgia railroad employees and citizens of the Dunnavant and Leeds, Alabama, well before it first appeared in print in 1930. This is 43 years after the alleged event. It is plausible that the tradition could date from the event itself.

The central hypothesis of the Alabama story is “The legendary John Henry was a steel driver who raced a steam drill at Dunnavant, Alabama, in 1887.” A network of testimony and facts supports the Alabama hypothesis, which is further supported by elements of the ballad not mentioned above. The Alabama story very coherent.

Accepting the Alabama hypothesis requires finding a plausible explanation for the facts underlying the Virginia story. That is easy enough.

The Virginia story hinges on a number of critical assumptions that are unsupported or poorly supported. If at least one of those assumptions were wrong then the facts underlying the Virginia story would not necessarily relate to the John Henry tradition and there would be nothing to explain. It is not only plausible but also very likely that some of those assumptions are wrong.

The central hypothesis of the Virginia story is “The legendary John Henry was a steel-driving convict laborer who died at Lewis Tunnel, Virginia, probably in 1871.” This hypothesis is consistent with a small number of facts that are augmented by many assumptions to generate the Virginia story.

Accepting the Virginia hypothesis requires finding a plausible explanation for the facts underlying the Alabama story. That is not easy.

It would require a number of very improbable coincidences. One would have to imagine as coincidence that informants from Utah, Michigan, and Jamaica named John Henry’s boss as Dabney/Dabner; that informants from Utah, Birmingham, and Michigan gave information pointing to Dunnavant, Alabama, and 1887 as the place and year of John Henry’s contest; that the man in charge of the construction of the railroad at Dunnavant was, in fact, named Dabney; that the years of construction were, in fact, 1887-88; that informants from Utah, Birmingham, and Michigan named Coosa and Oak Mountain Tunnels as John Henry’s places; that these tunnels were, in fact, bored at Dunnavant under the supervision of Captain Dabney in 1887-88; that informants from Utah and Leeds, Alabama, stated that John Henry and his boss were from Mississippi; that Captain Dabney did, in fact, live in Mississippi; and that his family owned slaves there, among which was at least one named Henry, who was the right age to have been a steel driver in 1887.

It is impossible to imagine that these are coincidences. If they were not coincidences and were not true, the alternative would be that an elaborate fictional story, the Alabama story, made the rounds for forty years and more, elements being recovered in 1927-32 and 1955 and being recoverable today in the living tradition at Dunnavant and Leeds. Where would this story originate if not in truth?

These considerations make the Alabama story far more probable than any alternative that I have been able to think of.

zak 822 - 12/5/2006

Just a sidebar to the discussion. You wrote about the C & W Railroad and the Lewis Tunnel.

My version of John Henry has the verse: "The Big Bend tunnel on the C & O Road,

Gonna be the death of me, Lawd, Lawd,
Gonna be the death of me."

I found my version in a book of songs in my junior high school library, in 1962 or 1963. I used to sing it for my father when we drove from Pittsburgh, Pennsylvania to Lynchburg, Virginia to visit our family there.