On the morning of October 11, 1874, a traveling Methodist preacher arrived at the Ballantine house by the long gravel road, carrying a Bible, a jar of apple preserves, and a letter from a worried mother in Louisville whose daughter worked there as a servant and had not written in nine months. The preacher knocked on the door for twenty minutes. When the maid finally opened it, he saw a girl behind her in the hallway, her eyes not following his face, her hands wrapped in cloth, her mouth sewn shut with what the preacher later swore in his testimony was fishing line. The maid smiled, took the preserves and the letter, closed the door, and never opened it again.
Six years later, in the summer of 1880, Sheriff Elias Hook of Bourbon County entered through those same doors with twelve armed men and found what he described in his official report as a system. Fourteen young women in the attic, nine in the cellar, and four in a sealed room behind the storeroom, its door without a handle on the inside. And in a stone building beside the creek, a ledger. A thick leather book in which the Ballantine family had recorded the names of 231 servant girls, their origins, their assigned duties, and their final disposition over the course of 41 years.
Most historians say this is all a hoax. They say the Ballantine estate burned down in 1881, and the records were lost. Most historians have not read the depositions taken in Paris, Kentucky, in September of that year, and they have not read what one of the rescued women, a Miss Hannah DeLuce, told the court before she was declared insane and sent to the Central Kentucky Asylum where she died. Because what she said was that the Ballantines were not from Kentucky.
They were not, she said, from anywhere. And the house, the land, the quartz vein running beneath the hill, and the four stone outbuildings arranged in a square around the main dwelling—all of it, she said, was already there when the first Ballantine arrived in 1839. Already there, waiting for them. The Ballantines arrived in Bourbon County, Kentucky, in the summer of 1839, a decade before the land they claimed to own was supposedly sold by the state.
There is no deed recorded in the Bourbon County courthouse transferring ownership, no record of payment, no record of any previous owner selling 1,200 acres along the Stoner River to anyone named Ballantine or to anyone else in the years before 1839. What was found in the margin of the county tax assessor’s book for 1840 was a handwritten note in a cursive that did not match the assessor’s own hand, reading only: “Ballantine property, assessed, paid in full, please do not disturb. ” No amount, no signature, no date beyond the year. And every tax assessor from 1840 to 1881 copied that note into their own books year after year, without visiting the property, surveying it, or asking a single question.
41 years of ownership of 1,200 acres in the heart of some of America’s most valuable horse-breeding country, and no official set foot on the land until Sheriff Hook arrived in June 1880. As far as anyone could later reconstruct, the founder was a man named Josiah Augustus Ballantine, aged somewhere between 40 and 60 on arrival, who claimed in a letter to a Lexington attorney that he had come from his country of origin through the old houses. The attorney, a man named Caleb Winstead, asked which country precisely. Josiah’s reply, which Winstead kept and which was later entered as evidence, was terse: “The land that was here before the water came the second time, and the houses that stood on the hills above it—you will not find them on your maps.
Nor do we list them on ours, and that is why we are still here. ” Winstead filed the letter, took his fee, and never contacted the Ballantines again. He died in 1852. His widow sold his papers to the Kentucky Historical Society in 1891.
The letter sat in box 14 of the Winstead correspondence until 1947, when a University of Kentucky graduate student named Theodore Marbury, preparing a thesis on antebellum Bluegrass plantations, found it. Marbury wrote three letters to the Historical Society requesting permission to publish the letter in full. The Society refused. Marbury then wrote to the Ballantine family directly, having discovered that their descendants still held an interest in the original land through a trust registered in Delaware.
The reply he received was a single unsigned page: “Mr. Marbury, stop this. There is no interest here in the servant question, no particular interest in the water question, and no interest whatsoever in what was before the water. Your thesis will not be accepted.
Your supervisor has been contacted. Find another topic. ” Marbury found another topic. He received his doctorate in 1949 for a study of early Shaker agricultural practices in Kentucky.
He died in 1988, and among his papers, left to his daughter, was a sealed envelope marked only: “Ballantine. Do not open unless I and my family are dead. ” The daughter never opened it. The family is still alive.
Josiah Ballantine married a woman named Clementine in 1841. There is no record of her maiden name, no record of her arrival in Kentucky, no record of a wedding, a marriage license, a witness, or a minister. She appears in family records only from 1842 onward as Mrs. Ballantine, wife of Clementine.
The 1850 census enumerator who walked the roads of Bourbon County that year recorded her age as 34 and her place of birth as “unknown, foreign. ” In the 1860 census, she was recorded as 34. In 1870, 34. In the 1880 census, taken only two months before the sheriff’s raid, her age was still 34, her birthplace still listed as “unknown, foreign.
” Four enumerators, all men in their forties, all with the same handwriting, though they were different men. One of them, a man named Reuben Craddock, left a note in his personal diary from June 1870, later donated to the Filson Historical Society in Louisville: “Visited the Ballantine home for the census. The lady received me on the porch and did not invite me inside. She has not changed since my father, who was the enumerator in 1850, described her as a young woman perhaps thirty.
I asked her age, and she gave me 34, which I wrote down for fear of writing otherwise. There are things in that house I did not see, and I am certain I did not see them, which is how I know they are there. ”
They had, officially, three children. Augustus Jr.
, born 1843; Prudence, born 1845; and Nathaniel, born 1849. None of them attended school. None appeared in any social register, church record, or militia list in Bourbon County. All three were listed as household members in every census through 1880.
All three were present when the sheriff arrived in June of that year, all three gave statements, and all three were tried and convicted, along with their parents, their housekeeper, and the four farm laborers who had worked the place since the 1850s. The three vanished from the records immediately after sentencing. The state of Kentucky claims they were hanged. The state of Kentucky cannot produce execution records.
It cannot produce burial permits. It cannot produce arrest warrants. It cannot produce trial transcripts, which, according to a clerk’s note in the Bourbon County criminal court records, were removed by order of the governor in September 1881 and never returned. The purpose of the servant girls was not house service in any ordinary sense.
The Ballantine estate had no guests. It held no social events. It received no visitors except the itinerant preacher who was turned away at the door, and the dry-goods delivery driver from Lexington who came twice a year to unload supplies at the end of the road and was never permitted inside the gate. There was no household in the usual sense—the family did not entertain, did not travel, did not receive anyone, did not go to church, did not go to town, and, as far as anyone later knew, did not even eat the food the household staff was asked to prepare.
Shipments arrived from Lexington twice a year. Loose flour, loose salt, loose coffee, a small amount of sugar, a large quantity of lamp oil. And one item listed on every shipping manifest from 1844 to 1880 that Bourbon County historians have been unable to identify. The manifests describe the item only as “the requested quantity of the requested substance, prepared by the usual party, delivered to the usual place.
” No name, no weight, no price—because it was paid for separately, in cash. The delivery driver who testified at trial, a man named Hollis Pennewell, said the requested substance was packed in sealed clay jars, four jars per delivery, and the jars were heavier than they should have been for their size. He said he once dropped one and it did not break, and the sound it made when it hit the ground was, in his words, like a cow lowing from far away, from under the earth, from a cow that had never been a cow. He said he never dropped another jar.
He said he was relieved when the deliveries stopped. The ledger, a thick leather book three inches thick found in the stone building by the creek, was divided into four columns. The first column was the servant’s name. The second was her place of origin, most often a small farm or town in Kentucky, Tennessee, Indiana, Ohio, or Pennsylvania, and in a few cases Ireland, Scotland, or a region described only as “the north country.
” The third column was her assigned duties, which made the record, in Sheriff Hook’s phrase, a document no person of ordinary education should be asked to read. The duties listed were not real duties—not cleaning, cooking, washing, or sewing—but things like “first watch of the second basin,” “naming at the quarter hour,” “opening the door,” “listening at the stone,” “feeding what is beneath the apron,” “breathing for the east hour,” and “receiving what returns. ” The fourth column was the final disposition, which in every case, in 230 of 231 rows, was the same four words: “returned to the house. ” There was only one different entry.
The last one, dated June 2, 1880, four days before the sheriff arrived, read: “The house would not accept her. She remains among us. We do not know what to do. ” The name on that entry was Hannah DeLuce—the woman who later told everything to the court before being declared insane and sent to the asylum where she died.
The ledger was not lost in the 1881 fire, as the common account claims. It was photographed. The photography was not high quality—photography in rural Kentucky in 1880 was not common—but it was clear enough that 18 pages of it survive in the collection of a private library in Frankfort, closed to the public since 1962 but permitting academic access by appointment. These pages were published in a limited academic study by the Vanderbilt University folklorist Dr.
Virgil Hollenbeck, published in 1954 under the title “Servitude and Its Reflection in Four Antebellum Bluegrass Farms. ” Two hundred copies were printed. Hollenbeck died in 1961, the coroner attributing his death to heart failure from stress—though Hollenbeck was 51 years old with no cardiac history, and his wife said he had received a letter a week before his death from a party who refused to identify themselves, after which he burned all his notes from 1952 onward, two years before the study was even published. His wife kept one copy.
She hid it in a hatbox in the attic of her mother’s house in Shelbyville, where it remained until the box was sold at public auction in 2003, bought by a librarian from Berea College for four dollars, who recognized its value and donated it to the Berea Appalachian Studies Collection, where it is now kept in a locked cabinet under the index number BAS MO 412. The first recorded duty, “first watch of the second basin,” refers to a stone basin about four feet in diameter, set into the floor of the eastern of the four stone outbuildings and lined with lead. When the sheriff’s men broke in on June 7, 1880, it contained approximately eighteen inches of a clear liquid that was not water. The deputy who entered the room first, a man named Pearle Chastain, dipped his finger into the liquid and withdrew it, and later said his finger came out dry.
The liquid moved like water, settled like water, reflected light like water, but transferred to nothing it touched. One of the servant girls assigned to clean that basin later testified before her removal to the asylum that the basin was called the second basin because there was a first basin elsewhere, which the family did not allow the servants to see, located, she believed, beneath the main house in a room reachable only by a hidden stair behind the storeroom. The first watch required sitting beside the second basin for six hours, from ten at night to four in the morning, speaking into the liquid every fifteen minutes a short phrase in a language the servant did not know, which Mrs. Ballantine taught her syllable by syllable, telling her it represented the proper names of what was beneath the first basin.
The servant who testified could still recite the phrase at the time of her testimony. The court stenographer recorded it phonetically. That audio copy survives in a portion of the court record not destroyed during the governor’s purge of 1881—specifically, a secondary copy made by a Lexington journalist named Hilbert Dowd, who was present at the session and published excerpts in the Lexington Transcript of August 14, 1881, before the newspaper’s offices burned on August 19, 1881, in a fire the fire marshal deemed of unknown origin, possibly arson, perpetrator never identified. Dowd was in the office.
Dowd survived. Dowd moved to Oregon within a month and never wrote another word about Kentucky. He died in Portland in 1922. The phrase, as the servant recited it phonetically and as Dowd preserved it, was: “Eka sheth balnuum kishareth kishareth entam entam um are no.
” The first words resemble those of an extinct river language, specifically Akkadian, and the last three resemble mangled English, as though someone had tried to speak English and only partly succeeded. The servant girls were taught these words syllable by syllable by a woman who never aged, speaking the names of something that lived beneath the first basin in a room beneath a house that was already standing when the Ballantines arrived. The second duty, “naming at the quarter hour,” was performed by another servant, usually newly arrived, who sat in a small room on the second floor of the main house, above what the household called the storeroom, which Hannah DeLuce testified was in fact the entrance to the stair down to the first basin. The naming servant held a small bell, about the size of a schoolroom hand bell, and every quarter hour from six in the evening until six in the morning she rang the bell three times and then spoke a name aloud, louder than the bell, from a list fixed to the wall beside her chair.
The list contained 47 names. They were not the names of the Ballantine family, nor of the servants, nor of anyone recorded in Bourbon County records or in any Kentucky census. Hollenbeck’s study, after textual comparison with surviving Hittite and Hurrian ritual tablets held at the Oriental Institute of the University of Chicago, suggests those names are the names of those the tablets call “those who hear through the stone,” who answer when their names are spoken in the correct order and at the correct time by a mouth that does not understand what it says. The naming servant did not understand.
That was the point. She learned pronunciation only, not meaning. And she was warned, as Hannah DeLuce said, that if she ever asked the meaning of those names, or if she began to understand them, she would be transferred to the fourth duty—the one performed in the sealed room behind the storeroom, from which, Hannah said, no servant ever returned the same. No servant was killed.
They were not, for the most part, physically harmed in any way that left marks on their bodies. According to the burial records, the servants were returned to the house. What Hollenbeck’s study argues, based on the testimony of the only rescued servant who could still speak clearly about her experience, a woman named Dovie Blackman, found alive and sane in an outbuilding, who gave a 30-page statement before being sent, like Hannah, to the asylum where she miraculously survived until 1931, is that the servants were returned to the house as something other than themselves. Dovie said: “The house is what took them.
It was not about the Ballantines, it was about the house, its structure, its land, the quartz vein under the hill, the four outbuildings, the first and second basins, the stone, the apron, the east hour. ” The house would consume what the servant was, then return a copy that performed her household duties, answered to her name, wrote letters to her family when required, and met the Ballantines’ needs—no longer, in any sense, as Dovie made clear to the court, a person. The Ballantines were feeding the house. That was their job.
That was why they were brought to Kentucky in 1839. Josiah’s letter to Caleb Winstead mentioned that his family came from the land that existed before the water came the second time because the Ballantines were not the owners of the property; they were merely employees. They were, as Dovie said, the house’s caretakers, and the servant girls were the fuel. The third duty, “opening the door,” was performed in an outbuilding north of the main house containing a wooden frame resembling a loom, but which on close examination by Sheriff Hook’s deputies proved to contain no threads, no shuttle, no cloth, and no weaving mechanism.
Instead it held a series of iron hooks arranged in a precise pattern around a vertical frame. The servant assigned to open the door stood in the center of this frame for four hours every night, arms raised, reciting from memory a sequence of numbers descending from a very large number in the tens of thousands down to zero over the four hours. The count began at a different number each night, set by Mrs. Ballantine at sunset.
Hannah DeLuce testified that the numbers, she believed, represented a measurement of distance, and that this distance was shrinking night after night over the years, and that the servants in the house joked, if that was the word, that when the count reached zero on a night it was supposed to reach zero, rather than a higher number, something would arrive. Hannah said she did not know what would arrive. She said Dovie thought she knew, but Dovie never said. What Dovie did say, in a separate portion of her 30-page statement quoted at length by Hollenbeck in his fourth chapter, was that the open weaving frame was not a frame for weaving cloth but a frame for weaving something else.
She said that when the servant stood in the center with arms raised and counted downward, something in the air around her—something invisible to the naked eye but felt on the skin of the throat and the backs of the hands—was drawn through the frame from side to side, slowly and patiently, thread by thread, like a skilled weaver. She said she performed this duty eleven times over the two years she spent at the property, and on the sixth time she opened her eyes during the count, which she was forbidden to do, and saw, for a fraction of a second, what the frame was drawing through itself. She refused to describe it. She told the court that if she described it, the court would stop writing, the stenographer’s hand would stop, the paper would refuse the ink, and even if that did not happen, which she doubted, the words themselves would not remain on the page.
She said she had tried to write it down for herself in a notebook hidden under a floorboard in her servant’s room, and the next morning the page was blank—not torn, not scratched, blank—as if the writing had been pulled from the paper during the night. She said after that she stopped trying to describe it, stopped looking at it, and stopped counting the numbers aloud when her turn came, only moving her lips—which Mrs. Ballantine noticed on the ninth time, after which Dovie was reassigned to the fourth duty, and that was how she ended up in the sealed room, and that was how she became the only servant, along with Hannah, to survive the fourth duty without losing the power of speech. There is another detail concerning the third duty that Hollenbeck’s study mentions, and it matters because it shows the system was not a local phenomenon but a continental one.
The number Mrs. Ballantine gave the servant responsible for keeping the door open each night was not random. Hollenbeck determined, by comparing the initial numbers Dovie remembered before her reassignment with archival weather records and calendar data, that the number matched the distance in a specific unit of measurement from the Ballantine property to a point in the Gulf of Mexico, about 80 miles southwest of New Orleans, which was moving northward by a small, constant amount every night. As the point moved, the initial number shrank.
Over the 41 years of the property’s operation, the distance had decreased, by Hollenbeck’s calculations, by approximately 14 miles. The point was approaching Kentucky at a rate of 1,300 feet per year. The system was designed to track the descending distance of the point, so that something would be ready when it arrived. The point did not arrive during the Ballantine operation.
The estate burned in 1881. By the calculations, the point now lies approximately 190 miles from the site of the main Ballantine house. It lies somewhere in central Mississippi. It is still moving north.
The fourth duty of the seven, called “listening at the stone,” was performed in the smallest of the four outbuildings, west of the main house, with no furniture, no windows, no interior lighting, and a floor of flat gray stone laid in a precise circular pattern around a central slab roughly three feet square. The listening servant was placed in this building at midnight, barefoot, instructed to stand on the central slab for exactly forty minutes and then report what she heard. Dovie described this duty in her statement: she heard nothing for the first ten minutes except her own breath. Then, she said, the breath stopped being her breath.
The breathing continued, but it no longer came from her chest; it came from the slab beneath her feet. It was not loud, not threatening, but steady. Dovie described it as the breathing of a vast creature in deep sleep. When she reported this to Mrs.
Ballantine after her first listen, Mrs. Ballantine nodded and said: “Good, he is still asleep. You will listen again tomorrow and the next day, and you will tell me if the breathing has changed. ” Dovie listened for fourteen nights before her reassignment, and the breathing did not change in all those nights.
She wondered later, in the asylum, whether it had changed since. The fifth duty was called “feeding what is beneath the apron. ” The term “apron” has caused the most confusion among later researchers, as no architectural sense of the word appears anywhere in the property’s records, not even in the servants’ own descriptions. When elderly Bourbon County residents in the 1950s explained the term to Hollenbeck’s informants, they described the apron as a geological rather than architectural feature—a kind of ledge or stone shelf extending beneath the main house at a depth of about 40 feet, opening below into a cavity whose dimensions no one had measured, and no one had dared to try.
The feeding servant was given a wooden tray at exactly ten o’clock at night on her duty night. The tray held nine small cups, each containing a precisely measured amount of the substance Hollis Pennewell brought twice a year from Lexington—the substance never named on the shipping manifest, which the delivery driver described as sounding like a cow lowing from underground when he once dropped a jar. The servant carried the tray to a trapdoor in the storeroom floor, lifted it, descended a wooden ladder into a stone room beneath the house, walked across the room to a point at the far wall where the apron began, and placed the nine cups along the edge of the apron in a precise pattern. She was told not to look over the edge.
She was told to turn and come back without turning her head. She was told that if she looked, the thing beneath the apron would see her looking and would remember her face, and would come for her eventually. Not that night, not that year, but eventually. And no family, no marriage, no move to another state would prevent what would happen when it came.
The servants obeyed. All of them. Dovie said: “Except for one servant, whose name I do not remember, and whose name was crossed out of the ledger, who looked over the edge in 1869 or 1870, and was returned to the house that night as a replacement, and who, three weeks later, in broad daylight, while performing her duties, simply vanished. She did not die, she did not disappear, she ceased to exist.
Her body remained for a moment where it had been standing, then it was gone. The clothes she was wearing fell to the floor empty. Mrs. Ballantine entered the room, picked up the clothes, folded them, and said only: ‘There is nothing more we can do for her.
‘”
The sixth duty was called “breath of the east hour,” and it was the only duty performed outside the stone outbuildings and outside the main house. It was performed on the east porch, facing the first light of dawn, from four-thirty in the morning until sunrise, every day without exception, by whichever servant Mrs. Ballantine designated at the end of the previous night. The breathing servant stood on the porch, facing east, and breathed.
That was all. She had to breathe deeply and regularly, counting ten breaths, then pausing for one breath, then another ten, for approximately ninety minutes or until the sun rose. She was told her breath was being used, but she was never told for what. Hannah performed this duty 79 times during her year and three months at the estate.
She said it was the easiest duty and the most terrifying. She said on cold mornings she could see her breath leave her mouth, and it did not rise upward as breath normally does—it went east in a thin straight line, like smoke drawn through a pipe, into the dim light of the horizon. When she tried to turn her head to break that line, she found she could not. Her neck did not respond until the ninety minutes were up and the sun was fully above the horizon, at which point her neck relaxed and she would not think of the east hour again until the next evening, when Mrs.
Ballantine read her the next day’s duties by lamplight. The seventh duty, which sits between the sixth and the fourth in the ledger’s order, is the one Hollenbeck could not decipher at all, and which no living servant testified to. Its name in the ledger, with only an English translation, was: “keeping the schedule between the hours when no servant is awake. ” Hollenbeck believed this duty was performed by the house itself.
He believed the house, over its 41 years of operation, had adopted, if that was the word, its own shift—performing its own work in a language the servants were not required to speak during the hours they slept. He believed, in other words, that the house was not merely consuming servants but learning from them in some way. Learning what? Hollenbeck did not say.
He died before finishing the chapter. His wife wrote in a note appended to the Berea copy of the study that her husband told her in the week before his death, over breakfast, that he had come to believe the seventh duty of the house was training. Training for the day when the house would need no servants at all. Training for the day when the water would come a third time and the system would need no human intermediary.
And the house, having consumed enough people over the decades, would be able to speak the names, ring the bells, count down the numbers, and feed the apron entirely on its own. That day, he told his wife, was closer than anyone outside the working families imagined. She wrote that down. She died in 1973.
The note is in the Berea file. Between 1831 and 1859, a period of 28 years, according to a careful review of local tax records, census anomalies, and local newspaper archives, at least 41 estates were established in the eastern and central United States under conditions identical or nearly identical to the Ballantine property. A founding couple arriving from an unspecified origin, no registered deed. A marginal note in a tax book directing the assessor not to disturb the property.
An estate ranging from 800 to 1,500 acres. A main house that, according to local tradition, was already standing when the founders arrived. Four stone outbuildings arranged in a square around the main dwelling. A quartz vein running beneath the hill.
A pattern of household staff who arrived and never left. And a founding wife whose age never changed across census records. 41 estates in Kentucky, Virginia, North Carolina, Tennessee, Missouri, Ohio, Pennsylvania, Maryland, Delaware, upstate New York, and three in Massachusetts. Of these 41 estates, 29 burned.
In 1879 and 1888, a wave of fires broke out that contemporary fire investigators attributed to various causes—arson by displaced farmworkers, lightning, kitchen accidents, anarchist agitation—anything other than the single pattern linking them all: that in every case, in the months before the fire, a local constable, visiting preacher, or curious journalist had begun asking questions about the household staff. The remaining 12 estates did not burn. They remain, to this day, owned by trusts registered in Delaware in the names of descendants of the original founding couples. These trusts hold properties valued in the billions of dollars.
The descendants’ names do not appear in social registers. They do not run for office. They do not give interviews. In the few cases where a journalist approached, the descendants responded with legal threats, financial offers, and in two documented cases from the 1970s, the journalist’s sudden and unexplained transfer to a different beat.
A different continent, by editors who could not later explain why they had made that decision. Hollenbeck’s study names three of the 12 surviving estates. The names are not repeated here, because the only time a researcher published those names, a man named Linwood Averitt, who published an article in 1973 in a small Appalachian studies journal, all copies of that issue were withdrawn within six weeks of publication. Every library copy was somehow replaced with a different issue bearing the same volume and number but containing a different editorial.
Averitt himself was found dead in a Charleston, West Virginia hotel room in April 1974, attributed by the coroner to an accidental overdose of a drug not known to be in his possession. His widow said he was in Charleston to meet a source. She said the source was descended from one of the three families named. She died of cancer in 1979.
No copies of Averitt’s original article survive. The journal itself ceased publication in 1976. Its editor, Perry Ganshoff, retired to Florida and refused to discuss Averitt or the withdrawn issue. When a graduate student visited him in 1992, Ganshoff, then in his eighties, said: “The important Appalachian studies journals were the ones that were allowed to keep publishing.
You are talking to me because I understood that. Linwood did not. Linwood is dead. Please leave.
”
The fourth duty was called “receiving what returns,” and it was the heart of everything. It was performed in the sealed room behind the storeroom, the room with no inner handle. At the start of each lunar cycle, one servant chosen by Mrs. Ballantine was placed in this room with a candle, a chair, a basin of water, and a long iron key.
The key fit no lock in the house. It was placed on the floor at the servant’s feet. The servant was instructed to sit on the chair, light the candle, and wait. Hannah DeLuce performed this duty three times.
She was the only servant, besides Dovie Blackman, who performed it more than once and remained able to speak afterward. Dovie performed it four times. Each time, Hannah said, the room grew colder during the first hour, then warmed, then grew hot enough that the candle began to melt from its base rather than its top, then cooled again. And then, at a moment Hannah could not predict and could not mark in any other way, the iron key moved—not slid, not fell—but moved across the floor from where it had been placed at her feet to a position about three feet in front of the door, where it stopped.
Hannah said that at that moment she had to pick up the key, approach the door, and strike it three times hard in a specific rhythm Mrs. Ballantine had taught her. Then the door was opened from the outside. Mrs.
Ballantine stood there. Behind her in the hallway was another servant—a servant Hannah had never seen before, a servant, Hannah said, wearing the same dress the previous servant who had entered the sealed room was wearing, with the same hair color, the same height, the same build, answering to the previous servant’s name, not recognizing Hannah, not seeming to remember she had been her friend, looking at Hannah with eyes, Hannah said in her final testimony, that were the eyes of the house looking through a face that had been her face. The servants were being replaced. The house was replacing the servants.
The servants who entered the sealed room entered as themselves and came out as something the house had made to resemble them. Hannah said the servants—the originals, Hannah and Dovie, and all the 230 copies who came before—were preserved. Where? She did not know.
In what form? She did not know. For what purpose? She had an idea.
She did not share the idea in court. She wrote it down. The paper she wrote it on was placed in an envelope. The envelope was sealed.
The envelope was handed to the court clerk. The envelope was removed by the governor’s agents along with the trial records in September 1881. The envelope was never found. There is a secondary account of what Hannah wrote in that envelope, from the court clerk himself.
A man named Ambrose Trumbull, who kept a private diary of his years serving the Bourbon County court, and whose great-granddaughter found the diary in a trunk in a Lexington attic in 2004. Trumbull wrote in an entry dated August 20, 1881, that before placing the envelope in the clerk’s safe, he had read what Hannah wrote—a single page in a trembling hand, in words not all English, but which he understood as if dictated rather than read. He wrote that what Hannah said in her note was that the servants who entered the sealed room and were replaced were held after their replacement in a state that was neither death nor life, in a place neither in the house nor beneath it, and that their function in that state was to speak the names continuously, forever, in the language they had learned syllable by syllable. In other words, the servants became part of the system that consumed them.
They did not die. They did not become ghosts. They became, as Trumbull wrote, quoting Hannah, “voices in the stone”—not stone voices, but voices carried by the stone, and which must be carried, because what sleeps beneath the stone must not be allowed to sleep without being spoken to. The replacements the family returned to the house were, in this account, hollow forms built by the house from the house’s own materials, performing the surface functions of the servants they replaced, while the real servants worked invisibly, constantly, beneath the floor, speaking into a silence that could not be allowed to become a real silence.
Trumbull wrote that after reading the note, he could not sleep for a week, that he never again went to the courthouse basement alone where the clerk’s safe was kept, that when the governor’s agents came in September to collect the trial records, he handed over the envelope with his eyes closed, that he resigned from the clerk’s office in October, became a dry goods merchant, and lived another 41 years without setting foot in a courthouse. His great-granddaughter donated the diary to the Filson Museum. It is accessible by appointment. The relevant entry is in volume four, pages 112 to 114.
Hannah died in the asylum in 1884. On her deathbed, according to a nurse who later gave a statement to a Louisville paper, Hannah said: “They are poured into the water. Those who came the second time. Those who are still coming.
Those who are still coming. ”
In the final chapter of Hollenbeck’s study—the chapter his wife said killed him—he argued that the Ballantine estate, and the forty other estates belonging to its network, were not merely houses where strange events occurred in the nineteenth century. They were, in Hollenbeck’s view, maintenance stations. This term, “maintenance stations,” comes from a passage in the Hollenbeck papers held at Vanderbilt University, which quotes a letter sent from an occupant of one of the surviving estates to an occupant of another, dated 1908, intercepted and photographed by a predecessor agency of the FBI during a separate, chaotic surveillance operation, and never properly indexed.
The letter refers to the writer’s own house as “a maintenance station on the eastern line, per the second founding. ”
The second founding. Hollenbeck argued there was a first founding—pre-Columbian—which was, in his phrase, the civilization referred to in Mesoamerican manuscripts as the Fourth Sun, and in the Sumerian King List under the names of kingdoms whose durations exceeded any plausible historical chronology, and which Western archaeology had agreed to ignore. Hollenbeck believed that civilization ended in a flood—not the biblical flood, but a later flood, one that erased most of what was built, leaving buried beneath the hills of what is now the eastern United States the remnants of its infrastructure.
Houses, in some cases stone outbuildings, quartz veins that might not be natural, and a system requiring maintenance by hereditary servants, passed down somehow from the lost civilization to our own, tasked with keeping the system running until, as Josiah Ballantine said, the water came a third time. The Ballantine property burned on the night of October 3, 1881, three weeks after the governor’s agents removed the trial records. The fire, according to the Bourbon County fire marshal, began in the kitchen. The fire, according to the testimony of a tenant farmer whose cabin stood a quarter mile to the east, began in all four stone outbuildings at once, which was impossible unless the fires were deliberate and coordinated.
The main house burned to its foundations. The stone outbuildings did not burn. They cracked. They split.
They settled into the ground. Sheriff Hook, who visited the site the next morning, said the stone outbuildings looked, in his words, as if something beneath them had pulled them down from inside, like a mouth closing around them. Hook resigned in November 1881 and moved to Oregon, where he died in 1902 of what his son described in a letter as “the thing he said would come for him, and which came. ” The letter is in the Oregon Historical Society.
The son, Elias Hook Jr. , died in 1947. His son, Elias III, died in 1988. Maria Hook Devine, a woman in her sixties of the Hook line, refuses all media inquiries regarding the Ballantine case, refuses to confirm or deny whether she possesses any family papers relating to it, and lives in a house in rural Oregon surrounded, according to Google Earth, by four stone outbuildings arranged in a square.
The house sits on a hill. The outbuildings are visible. They are stone. They are square.
There is a final testimony, from a servant named Rosalind Bence, assigned to the Ballantine house in the spring of 1877 by her uncle, a small farmer in Fayette County, who within three months of her arrival understood what was happening in the house and made a decision no other servant had made in the 41 years of its operation: she decided to leave. She did not announce the decision. She did not pack a bag. She did not say goodbye to anyone.
She waited for the night she was not assigned any duty by Mrs. Ballantine—which happened only on the night of the new moon—and walked out the front door of the main house at about 11:00 at night, carrying nothing, wearing only her uniform and shoes, walked through the dark fields, through the woods, down to the Stoner Creek, where she walked upstream for three miles because Dovie had told her running water would confuse the house. She left the creek near a tenant farm, whose owner, a man named Prentice Hollingsworth, took her in without asking questions, fed her, hid her in a cellar for a week, and when he saw it was safe, moved her by wagon at night to a rail station in Carlisle, where she took a train to Cincinnati, then to Pittsburgh, then to Philadelphia, where she lived the rest of her life under the name Rosalind Mercer—her mother’s maiden name—working as a seamstress, marrying a shoe cutter, raising three children, and dying in 1932 at the age of 73. She never told her children about Kentucky.
She never told her husband. She wrote one letter in her own hand on the morning of her death, placed it in an envelope, and gave it to her eldest daughter with instructions not to open it until 40 years had passed. The daughter kept the letter. She died in 1968.
Her son opened the letter in 1972, exactly 40 years after Rosalind’s death, as instructed. The letter read: “To whoever opens this letter, I am Rosalind Bence Mercer. I was born in 1859 in Fayette County, Kentucky. I spent three months of my seventeenth year as a servant in the Ballantine house in Bourbon County, later famous for a reason I will not mention here, because mentioning it would bring back painful memories.
I escaped on the night of the new moon in July 1877. I walked along the Stoner Creek. I write this letter because I want someone to know, long after my death, three things. First, the house is real.
Not a metaphor. Not a rural nineteenth-century ghost story. The house is a self-contained entity. It has a mouth.
It has a hunger. It is older than the family that inhabited it, and the family that inhabited it is older than America. Both descend from the same place, which is not a country on the map. Second, the house is not alone.
There are other houses. I do not know how many. I know of two others for certain, because the Ballantines received visitors from them twice during the three months I spent in the house. And I heard through the storeroom door, which was less tight than the family believed, talk of the eastern line and the western line and the Gulf line, and of keeping to the schedule.
The schedule has to do with something approaching. The schedule has to do with the water, as the Ballantines call it, which is not really water. Third, if you are reading this, and if you came to this letter through your research into the Ballantine case or any of the cases like it, I want you to understand that your research has already been noticed. Interest in the families who inhabited the house is not something granted lightly or wasted on the harmless.
If you have come this far, you have become visible. I advise you to stop. I advise you to burn this letter and publish nothing. I advise you not to tell anyone what you have read.
The families concerned cannot reach everyone, but they can reach anyone they choose. And the choice is not random. They choose those who understand. Do not understand.
Pretend you did not read this letter. Pretend you did not find these records. Live a long quiet life as I did, and die in your own bed as I will. The house will come for the world in the end.
It is not something you or I can stop. Spare yourself the attention and the effort. ”
There is an additional detail from Rosalind’s letter that was not published in the 1989 quarterly because the journal’s editor, a man named Phineas Oakhurst, told her grandson it was too unbelievable and would damage the article’s credibility. The detail is this: Rosalind wrote that on the night she climbed out of the Stoner Creek, she looked back once from the hill overlooking the Ballantine property before descending to Prentice Hollingsworth’s farm, and she saw the main house from a distance of about half a mile in the dark, with the new moon low on the horizon, and she noticed that the house was not sitting on its foundations the way houses sit on foundations.
She wrote that the house was floating. Not high, only a few inches, but it was suspended above its stone foundations, with a thin band of dark air visible between the base of the house and the stone. And she wrote that the four outbuildings were doing the same. All of them floated on the new moon night in perfect stillness.
She wrote that she never told anyone because she did not expect to be believed. She wrote it in the letter because she said if the person who opened it decades later had come to it through the right research, they would not need convincing. They would already know. They would find other accounts from other estates where eyewitnesses had seen the same thing.
The houses rose. And the families who inhabited them held them down. That was what those families did. The servants fed the house.
The family held the house down. And the house, held and fed, stayed in place above the stone chamber whose contents must not be allowed to rise. The system was a weight. The servants were the counterweight.
When the counterweight ran out, the house would rise too high. And what the house was holding down would rise. Rosalind’s grandson, who published the portion quoted above in the regional historical quarterly in 1989, suffered what the Philadelphia Inquirer described as a sudden and unexplained fire in his Chestnut Hill home in February 1991. He survived.
His papers did not. The regional historical quarterly ceased publication in 1992. The grandson lives now, as of 2024, in a retirement community in Boca Raton. He refuses all media inquiries.
He has told his family the letter should never be mentioned, nor the Ballantine name, and that if a journalist or researcher contacts them, they should say they have never heard of it. In 2019, a researcher at the University of Louisville named Dr. Imogene Fairweather, working on a separate dissertation on antebellum household economies, requested access to the Bourbon County criminal records for 1881. She was told the records were unavailable for research use.
She filed a Kentucky Open Records request. It was denied, citing an exemption for materials whose disclosure would constitute an unwarranted invasion of personal privacy. She appealed. The appeal was denied.
She filed suit in Franklin Circuit Court. The suit was dismissed in September 2020 for lack of standing, on the grounds that the records in question did not pertain to Dr. Fairweather or any member of her family. Undeterred, Dr.
Fairweather filed a federal FOIA request for the records of the FBI’s predecessor agency, which had intercepted the 1908 letter mentioned earlier. The FOIA response, dated January 2021, confirmed the agency had held those records at one time, but informed Dr. Fairweather they had been destroyed according to standard retention schedules in 1973. Dr.
Fairweather filed a mandatory declassification review request for the same materials. The review response, dated June 2022, informed her the records had been transferred to another agency in 1973, and that agency had no record of receiving them. She filed a request with the National Archives. The National Archives had no record.
Dr. Fairweather, 44 years old, tenured, in excellent health, published a short article about her experience in the Journal of Southern History in the fall of 2022. The article was two pages. It did not mention the Ballantine family by name, referring only to a nineteenth-century estate case in central Kentucky involving significant archival gaps.
Dr. Fairweather has published nothing on the subject since. When contacted in the spring of 2024 about this story, her university email returned an automated reply stating she was on research leave and would return in the fall. In the fall of 2024, the automated reply continued.
In the spring of 2025, the automated reply continued. In the fall of 2025, Dr. Fairweather’s name was removed from the University of Louisville faculty directory. The university’s history department, when asked, declined to comment on whether Dr.
Fairweather still worked there. Her current whereabouts are unknown. It is to be hoped she took the advice offered to her, in whatever form it came, and decided to live quietly for the rest of her life. It is to be hoped the house, when it wakes, does not reach her first—or any of us first.
Though that hope is one the schedule does not appear to respect.
