The Bat Creek Stone
An inscribed tablet from a Tennessee burial mound, promoted since the 1970s as proof that Hebrew speakers reached America before Columbus. The letters were copied out of a Freemasonry dictionary.
Why this verdict
Debunked by its source: the inscription matches an illustration of a paleo-Hebrew phrase printed in a Masonic reference work reissued repeatedly through the second half of the nineteenth century, and available to the excavator who reported finding it in 1889.
- Period
- Region
- Reading time
- Sources
01The claim
That the stone tablet reported by John W. Emmert from Tipton Mound 3 in Loudon County, Tennessee, in February 1889 carries a first- or second-century paleo-Hebrew inscription, and therefore documents pre-Columbian contact between the Old World and North America — a contact the archaeological establishment is said to have refused to acknowledge.
02Where it came from
Emmert was excavating for Cyrus Thomas's Bureau of Ethnology mound survey, the project that was in the process of demolishing the 'Mound Builder' myth by showing that Native Americans had built the mounds. Thomas read the marks as Cherokee syllabary and used them to support that argument. The Hebrew reading came eighty years later, from Cyrus Gordon, a scholar of Near Eastern languages and a proponent of pre-Columbian transoceanic contact, who argued the stone had been published upside down and read the inverted characters as ancient Hebrew. The economist J. Huston McCulloch developed the case at length from 1988.
03What the record shows
✕ DEBUNKED — Investigated and refuted. Here is what actually happened.
Thomas's own reading was already impossible: the Cherokee syllabary was invented in 1819, which would make a mound containing it far younger than the archaeology allows. That incongruity was noticed at the time and quietly set the stone aside.
The decisive work is Robert Mainfort and Mary Kwas's 2004 paper in American Antiquity. They matched the inscription to an illustration of the paleo-Hebrew phrase QDŠ LYHWH — 'holy to Yahweh' — published in a General History, Cyclopedia and Dictionary of Freemasonry, a volume reprinted extensively in the later nineteenth century and readily available. The correspondence is close enough that Kyle McCarter, a specialist in Northwest Semitic epigraphy at Johns Hopkins, concluded it was not coincidence but fraud, and that the stone has no place in the corpus of Hebrew inscriptions.
The supporting context also failed. Bracelets from the same mound, described by Emmert and Thomas as copper, were shown by Smithsonian analysis in 1970 to be heavily leaded yellow brass — an industrial alloy. Emmert excavated the mound alone. Neither the extensive Tellico Archaeological Project excavations in the Little Tennessee Valley nor any other work in the region has produced a single artefact indicating Old World contact.
A further detail sits badly with everyone: two vertical strokes appear on the stone in photographs taken after 1970 that are absent from the 1890 lithograph and the earliest photograph. Someone cut them while the object was in the national collection.
04What it cost
The claim persists because it is useful to two separate constituencies — pre-Columbian contact enthusiasts and religious traditions with a doctrinal stake in ancient Israelites in America — and because it is a small object with an unglamorous debunking. The stone belongs to the Smithsonian and has been displayed on loan, latterly at the Museum of the Cherokee Indian. Its real historical significance is what it was originally used for: a fabricated inscription inserted into the one project that was successfully proving Native Americans had built the mounds themselves.
05Connections
06Sources
Primary document · 1894
Report on the Mound Explorations of the Bureau of Ethnology
Government · 2004
The Bat Creek Stone Revisited (state archaeology copy)
Academic · 2004
The Bat Creek Stone Revisited: A Fraud Exposed
Academic · 1988
The Bat Creek Inscription: Cherokee or Hebrew? — the case for authenticity
Encyclopedia
Bat Creek Stone