The Myrtles Plantation: Counting Ghosts That Multiply in the Telling

The Myrtles Plantation, a handsome antebellum house near St Francisville, Louisiana, is routinely marketed as one of the most haunted homes in America, and the pitch comes with an unusually precise claim attached: that ten murders were committed on the property. The star of its resident mythology is Chloe, an enslaved woman in a green turban who, the story goes, was the mistress of the plantation’s owner, Judge Clark Woodruff. Caught eavesdropping, Woodruff had her ear cut off, which is why she wore the turban. In revenge, or in a bid to make herself indispensable by nursing the family back from an illness she herself induced, Chloe baked a birthday cake laced with oleander leaves. Woodruff’s wife and two of his children ate it and died. The other enslaved people, terrified of reprisal, dragged Chloe out and hanged her, then weighted her body and threw it in the river. Her turbaned figure is said to wander the house still.

It is a vivid, specific, terrible story. It is also, in almost every checkable particular, false — and the ways in which it is false are a small masterclass in how a haunted-house legend assembles itself from grief, guilt, and the simple commercial usefulness of a good ghost.

The kernel: a real house with a real and violent past

Begin with what is true, because the foundation is solid enough to make the embroidery believable. The house exists and is old. It was built around 1796 by General David Bradford, a lawyer who had fled to Spanish Louisiana after leading part of the Whiskey Rebellion in Pennsylvania. He called it Laurel Grove. His son-in-law, Clark Woodruff, did marry into the family and did live there; later the plantation passed to Ruffin Gray Stirling, who enlarged and renamed it the Myrtles in the 1830s. Like almost every plantation in the antebellum South, it was a site of enslavement, and enslavement is violence by definition — the suffering on that land was vast, continuous and real, whatever the ghost tour makes of it. St Francisville and the surrounding parish also saw genuine bloodshed during and after the Civil War, an era of guerrilla killings and reprisals across the Felicianas.

And there was, at the Myrtles, a real murder. In 1871, a man named William Drew Winter, who had married into the Stirling family and lived at the house, was shot by an unknown assailant who called him out onto the porch. He died of the wound. That killing is documented, a court and press matter rather than a legend, and it is the historical seed around which much of the house’s death-lore later crystallised. The Myrtles carries genuine darkness in its own right: enslavement, war, and one verifiable homicide. The question is how that became ten.

The fork: the murders that never happened, and the woman who never was

Here the record and the legend part company hard. Historians and researchers who have worked through the archives — parish records, censuses, the documented histories of the Woodruff and Stirling families — find no support for the central atrocity. Clark Woodruff’s wife, Sara, and their children did not, according to the surviving record, die of poisoning at all; the family’s deaths are better accounted for by the ordinary killers of the nineteenth-century South, yellow fever chief among them, which carried off staggering numbers along the lower Mississippi. Woodruff himself did not stay on to be haunted by remorse; he left the plantation and moved away.

As for Chloe: there is no documentary trace of an enslaved woman by that name at the Myrtles, no record of the ear-cutting, no record of the poisoned cake, and no record of a lynching by the other enslaved people. She appears to be a wholly legendary figure, a character rather than a person, and the specific business of the green turban has the tidy after-the-fact logic that invented details usually do — it explains a striking image (a woman in a headwrap) by supplying a mutilation to justify it. The “ten murders” figure fares no better. Serious tallies of documented deaths on the property come to something far smaller; the honest count of murders the record can actually stand behind is essentially one, William Winter’s, in 1871. Everything between one and ten is accumulation.

How the count grows: the anatomy of a multiplying legend

Watch the mechanism, because it is general. A house acquires a reputation for being haunted; a reputation for being haunted wants deaths to hang itself on; and deaths, unlike ghosts, can be counted, which makes them irresistibly quotable. “The most haunted house in America” is a boast. “The site of ten murders” is a statistic, and statistics travel further and sound more authoritative than adjectives. So the number is the part that inflates, because the number is the part that sells.

Each retelling has quiet permission to round up. A guerrilla killing somewhere in the parish during the war becomes a killing “at the Myrtles.” A death in the house from illness becomes a death by violence. Winter’s single documented murder anchors the tally at a real one, and then the imagination, needing the house to be as terrible as its atmosphere promises, supplies the other nine from the general reservoir of Southern historical horror — of which, tragically, there was always plenty to draw on. The oleander cake, meanwhile, is a portable folk motif that attaches itself to grand houses across the South and beyond: the enslaved poisoner taking revenge through the kitchen is an old and recurring anxiety of the slaveholding class, and the Myrtles simply gave it a name and a turban. None of the individual embellishments needs a liar behind it. A legend of this kind grows the way a snowball does, by rolling downhill through generations of people who each add a little and subtract nothing.

The photograph, and the machinery of proof

The modern Myrtles also has its evidentiary relic, the counterpart to Raynham Hall’s staircase plate. In 1992, an insurance photograph taken of the property is said to show a faint human shape standing in the gap between two buildings — instantly christened Chloe, and reproduced ever since as visual proof that the turbaned poisoner still walks. As with every such image, the shape is ambiguous, low-resolution and open to the ordinary explanations: a trick of light and shadow in a narrow gap, a smear or flaw, a chance arrangement that the eye, already primed by the story, resolves into a figure. The photograph did not create Chloe; the legend of Chloe created the readiness to see her, and the photograph merely gave that readiness something to point at. This is the sequence in almost every famous ghost picture — the story comes first and the image is recruited afterward, never the other way round.

The house leaned into all of it, and sensibly so from a business standpoint. Under later owners the Myrtles became a bed-and-breakfast and a tour operation, and the ghost story is the product: the poisoned cake, the murdered children, the ten deaths, the girl in the turban seen on the stairs. There is nothing especially sinister in a struggling old plantation house monetising its atmosphere; historic properties are expensive to keep standing, and a good ghost pays the roof bills. But the commercial incentive is exactly why the count never shrinks. No tour has ever gained a customer by announcing that the archives support one murder and no Chloe.

What it’s really about

Two things are really going on beneath the Myrtles legend, and they sit uneasily together. The first is the ordinary life-cycle of a ghost count: a memorable house, a single true horror, and a folk process that rounds one killing up to ten because a bigger number is a better story and nobody is checking. That part is almost comic in its mechanics, and it is the same process that gives every “most haunted” site its inflated body count.

The second thing is not comic at all. The legend of Chloe is a fiction, but it is a fiction stitched together out of a real and monstrous history. There were enslaved people at the Myrtles, and their suffering was total and largely undocumented, their names mostly lost. The green-turbaned poisoner is a story the slaveholding culture told about its own guilt and its own fear — the terror that the people it brutalised might one day answer through the kitchen — and it is telling that this invented, sensational Chloe is now the most famous “enslaved person” associated with the property, while the real men, women and children held there have no monument, no tour, and often no recorded name. The haunting, in that light, is a kind of substitution: a lurid, saleable ghost standing in front of a genuine mass of unremembered dead. When we count the Myrtles’ ten murders, we are counting a number the archive cannot support, in front of a history whose true toll no one ever bothered to write down. The most honest thing that can be said about the house is that it is haunted, but not in the way the tour describes, and by far more people than it names.

The one death the record can stand behind

It is worth staying a moment with William Winter, the single documented murder, because the way his real death has been folded into the legend shows the whole process in miniature. Winter was shot on the Myrtles’ porch in 1871 by a man who reportedly called him out of the house, and he died of the wound. Over the years the ghost tour elaborated the killing into something more theatrical: in the popular telling Winter, mortally wounded, staggered back inside and up the grand staircase, dying on the seventeenth step, and his dragging footsteps are said to climb those stairs still. The staircase detail is dramatic, memorable, and unsupported — a piece of stagecraft grafted onto a real corpse. That is the Myrtles method exactly. Take the one death the archive confirms, keep the name and the date because they lend the story its ring of fact, and then embroider the manner of dying into something a guide can point at on the stairs. The real Winter, a man genuinely murdered in his own home, becomes a prop. And around that one true, embellished death, the legend arranges its nine phantom others and its poisoner in the green turban, until the house’s advertised body count bears almost no relation to anything that can be found in a courthouse record.

For a ghost photograph whose story likewise arrived before the image, see The Brown Lady of Raynham Hall: The Most Famous Ghost Photo, and for another American house whose legend was built by the tourist trade, The Winchester Mystery House: The Widow and Her Endless Stairs.