The Grave Creek Stone

The Grave Creek Stone was discovered in 1838 during the excavation of the Grave Creek Mound, in Moundsville, West Virginia, on the Ohio River, about 10 miles south of Wheeling. The stone, an actual photograph of which appears in Figure 1 above, was a small inscribed sandstone disk, about 1 7/8" (4.8 cm) wide, and 1 1/2" (3.6 cm) high. The reverse side was uninscribed.

In 1838, the mound was reported to have been 69 feet high and 295 feet in diameter at the base, making it the largest of the Adena mounds. Today the mound is preserved in the Grave Creek Mound State Park. According to the Park website, the mound is currently believed to have been built between 250 BC and 150 BC. Figure 2 below shows the mound in its present condition. Note the relative size of the cars and house.

Fig. 2
The Grave Creek Mound today
Lifted from Home in West Virginia website.

The Cast and Wax Impression

In 1868, however, Davis made a plaster cast of the stone and deposited it in the Smithsonian Institution. In 1990, Donal Buchanan and I visited the Smithsonian's National Museum of Natural History in order to view the cast and to check out rumors that the Smithsonian had the original. The NMNH in fact has four casts of the stone, but no original.

(C) Smithsonian Institution

Fig. 3.
Cast of the Grave Creek Stone.
Smithsonian Institution Photograph No. 6768
(Catalogue No. 7252)

Two of the casts have catalog number NMNH 7252, one NMNH 138470, and one NMNH 325934. The darker of the two 7252s, which I call 7252 #1, is painted realistically and signed at the bottom of the front in pale white ink, "E.H. Davis W. Va." It is shown in Figure 3 above. The white marks under the figure at the bottom are just the remnants of Davis's signature, upside down. The second, lighter colored NMNH 7252 #2 and the two higher numbered casts appear to be derivative casts made from the above NMNH 7252 #1.

The National Anthropological Archives (MS 3146, E.H. Davis Collection) also has a wax impression of the stone that Davis made at the same time. It is unprotected and badly cracked, but confirms some details that do not show well on the cast. In particular, it clearly shows an X-shaped letter at the left end of the second line, and a lozenge- or diamond-shaped letter at the left end of the third line, that only partially "took" on the cast.

Our inspection confirmed the view of Charles Whittlesey, (1876) that the drawing by Capt. Seth Eastman, which appeared in Henry R. Schoolcraft's 1850 Indian Tribes of the United States, was the only reliable drawing of the stone, of the six depicted by Whittlesey. Unfortunately, many of the early scholars who made serious attempts to interpret the inscription, including Rafn, Jomard, and Bing-Lévy, worked from seriously inferior copies of the stone. This may at least in part account for the total divergence of the early interpretations, as to both language and content.

Fig. 4
Photocopy of Seth Eastman drawing of Grave Creek Stone
from Henry R. Schoolcraft Indian Tribes of the United States, 1850,
by way of Barnhart (1986).
(added 2/08)

A New Drawing

Below is my own attempt to draw the inscription, based on a tracing of the above photograph of the cast, supplemented by our inspection of the actual cast and wax impression.

Fig. 5.
The Grave Creek Inscription

It is not obvious from the inscription itself whether the lines are to be read from left to right or right to left, or perhaps even back and forth or "boustrophedon." It is not even obvious which end of the stone is the top. Many of the letters are invertible top for bottom, and two actually appear along with their own inverse. I have merely followed the universal convention of placing the figure at the bottom.

The horizontal guide lines are more shallow than the letters themselves, and are not part of the inscription. In Figure 5 below, I have eliminated them and separated the lines for clarity.

Fig. 6.
The Letters of the Grave Creek Inscription

The principal difference between my drawing and Eastman's is that in the D-like second letter from the right in the last line, Eastman has drawn the crossbar as a faint wavering line, which one might interpret as a stray scratch, while in fact it is straight and bold, and definitely part of the letter. I have drawn the vertical of the "D" to extend down below the crossbar to meet the loop, as Eastman did, since this appears to be what was intended. However, this continuation in fact is not very clear on either the cast or the wax impression. It is therefore not inconceivable that the vertical only extends down to the crossbar, as it appears in Whittlesey's Copy #2, from the American Pioneer, May 1843, and Copy #3, used by Jomard.

The Early Controversy

Historian Terry Barnhart has recently published an excellent overview (1986) of the 19th century controversy over the authenticity of the Grave Creek Stone. Schoolcraft, Wills de Hass, and J.P. MacLean supported its authenticity, while E.G. Squier, Davis, and M.C. Read regarded it as a forgery. Barnhart's most interesting observation is that in 1847, Squier had made much of the "singular omission" of any mention of the stone in Dr. James W. Clemens' first-hand, day-by day account of the excavation, which appeared in S.G. Morton's 1839 Crania Americana. In 1858, however, de Hass managed to produce the manuscript original of Clemens' account, and demonstrated that Morton had merely taken it upon himself to expurgate the stone's discovery from the published version. Dr. Clemens in fact recorded the inscribed stone on the day of its discovery.

In 1877, the Ohio State Archaeological Society appointed a committee to study the authenticity of the stone. Committee member M.C. Read published a strong denunciation, that included a widely cited experiment to determine if the inscription was alphabetic: He asked four persons who had no training in ancient inscriptions to fabricate twenty or more arbitrary "letters" composed of straight lines and combinations of straight lines. "The result, said Read, was that 'In every case an inscription was produced presenting as many indications of being alphabetical' as that on the Grave Creek stone. Accordingly, Read concluded that the Grave Creek inscription was 'just such a medley of characters as anyone would produce who undertook to invent an inscription to puzzle the curious.'" (Barnhart 1986 p. 120)

Committee member Rev. J.B. MacLean, on the other hand, published his own article in which he "did not hesitate to pronounce its authenticity as incontestable.... Regardless of who found the stone or whether it was discovered inside or outside the mound, all professed witnesses agreed it had come from the mound. To MacLean, this was the unassailable preposition." (Barnhart 1986 p. 122)

Barnhart's discussion stops short of the two influential articles by Charles Whittlesey, already mentioned above. The first, entitled "Archaeological Frauds," (1876) deals at length with the Grave Creek Stone. Whittlesey ominously cites Squier's finding that "Dr. Clemens, in his first account of the opening of the mound, makes no mention of this stone" (p. 5), but himself makes no mention of de Hass's correction of this misconception.

Whittlesey condemns the inscription itself on the very peculiar grounds that even "If the Grave Creek find was free from suspicion as to its integrity, it has undergone so many mutations from transcribers and translators that its value to ethnologists is gone." (1876, p. 5) Whittlesey himself had already demonstrated that most of the copies that had been made of the stone were unreliable. However, even though these inaccurate copies created considerable unfortunate confusion among scholars, they in no way altered the inscription itself or in any way lessened any ultimate value it might have to ethnologists.

It was also true that the many attempts at translation differed completely as to substance and even purported language and alphabet. However, this does not demonstrate that none of the proposed translations could be of any merit, but merely that at most one could be of any merit. Even if they all proved erroneous (which Whittlesey did not even begin to attempt to demonstrate in either of his articles), it still would not follow that there could never be a valid translation. Nor would they in any way hamper any future valid translation. In any event, the disagreement among the translations may have been more the fault of the bad copies the translators had to work with than of the inscription or of the translators themselves.

Whittlesey concludes that "The best authorities in the United States have condemned it during many years. The preponderances of proof as well as of probabilities are decidedly against it." (1876, p. 5) The very title of his article clearly classifies it as a fraud.

In his second, 1879 article, entitled "The Grave Creek Inscribed Stone," Whittlesey intriguingly backs off his earlier position that the stone itself is a clearcut hoax. Instead, he now merely insists that the inscription is not alphabetic: "I agree with Prof. Read that the characters on the stone, by whomsoever they were cut, are not alphabetical or phonetic. If they have any meaning and are not a mere jumble of characters they must be symbolic or picture writing. It is therefore of small consequence whether the stone is antique or modern, whether it is genuine or a fraud." (p. 66) He concludes, "If Professor Read and myself are right in our conclusions, that the figures are neither of the Runic, Phonician, Canaanite, Hebrew, Libyan, Celtic, or any other alphabet language, its importance has been greatly overrated." (p. 68)

After Whittlesey's two articles the Grave Creek Stone was generally dropped from serious consideration by archaeologists, except as a textbook example of an established hoax. It was so thoroughly discredited that they even lost track of its whereabouts.

Wills de Hass was appointed to head the Smithsonian Bureau of Ethnology's Mound Survey project, but was quickly replaced in favor of Cyrus Thomas. It is not unlikely that this had something to do with his favorable position toward the Grave Creek Stone. Thomas, on the other hand, "took a very strong, almost sarcastic, stance toward the Grave Creek Stone and Schoolcraft's support of it." (Williams 1991, p. 86)

The Bil Stumps Reading

A 1930 article in Science News Letter reported in all seriousness that one Andrew Price had at last "cleared up one of the greatest hoaxes in the record of American science," by translating the Grave Creek inscription into "good old West Virginian." (Davis 1930) According to the article, "That hoax, perpetrated by some unknown practical jokesmith, has stood triumphantly undetected for ninety odd years." The inscription, according to Price, reads, in plain English,

Price claims that this hoax was inspired by Charles Dickens' The Pickwick Papers, published just the year before the Grave Creek mound was opened. In Dickens' novel, Mr. Pickwick discovers a stone bearing the following cryptic inscription:

The Bil Stumps reading is an amusing spoof that has no true relation to the inscription. Even Williams (1991, p. 87) concurs in this assessment. The only hoax here was that the likes of Science News Letter fell for Price's "solution."

The Hough Stone

Figure 7.
The Hough Stone
Photo courtesy Robert B. Miller, Jr.
[Added 8/31/00]

In 1951, Philip R. Hough purchased for $1 an inscribed stone from an assemblage of unprovenienced Indian artifacts offered for sale by a gun dealer near Steubenville, Ohio. Hough sent a rubbing of the stone to the Smithsonian. He was informed that it closely resembled the Grave Creek stone, and might either be the original or a clever copy. Hough duly reported his find and its circumstances in a note in the Tennessee Archaeologist (1952).

In 1988, Hough's grandson Robert B. Miller, Jr., then of Richmond Va., sent a photograph of the Hough stone to the late Victor Moseley, then president of the Midwestern Epigraphic Society. Moseley passed the photo on to me, which is displayed in Figure 6 above.

A close comparison of the Hough stone to the cast shown in Figure 3 above reveals, unfortunately, that it is not the original Grave Creek Stone. Its guide lines are too straight, too evenly spaced, and too parallel. Furthermore, the vertical alignment of the letters does not match that on the cast. It is, however, definitely a deliberate modern copy of the Grave Creek stone, and not, like the Braxton Creek and Ohio County stones, a possible corroborating inscription in the same script. There is no evidence that whoever made it was attempting to pass it off as the genuine article, and Hough is to be commended for his straightforward reporting of the matter.

Fell

Modern interest in the stone has been revived, at least among amateurs like myself, by the late Barry Fell in his 1976 book America B.C. (1976a, p. 21, ch. 11). Fell offers yet another entirely different translation of the inscription, stating that the script is Iberian and the language Punic:

Fell provides the details of his translation in (1976b). In this article, he generally follows Eastman's drawing of the stone. However, it should be noted that he adds a short foot to the third letter from the left in the second line, and then reads it, together with its two neighbors, as a single letter (a Semitic shin). I see no evidence of a foot or any connection at the base apart from the guideline, which should not be taken as part of the inscription, and believe this group should instead be read as three separate letters. I have no idea how this change would alter sense of his reading.

Fell's translation is incorporated into an exhibit about the stone in the Delf Norona Museum at the Grave Creek Mound State Park. The replica of the stone on display there is not an actual cast of the original stone, but merely an artist's reconstruction based on Eastman's drawing.

Humbug?

The Grave Creek Stone plays the lead role in Chapter 4 of Stephen Williams' 1991 Fantastic Archaeology, entitled "The American Humbug: They'll Believe Almost Anything!" Williams initially has the stone first noticed by Schoolcraft five years after the excavation among some artifacts taken from the mound by its owner, Abelard Tomlinson. In his next paragraph, however, Williams indicates that Tomlinson actually exhibited the stone about two years after the excavation (p. 82). Either way, he warns, following Samuel F. Haven, that the claim that the stone came from the mound was greatly "weakened by the time that purportedly elapsed between its discovery and the announcement of the find" (p. 84). He makes no mention of de Hass's 1858 demonstration that Clemens in fact recorded it on the day it was found. Instead, he cautions the reader that "the problem of a time delay is a common thread running through a number of Fantastic Archaeology cases" (p. 85).

Williams places great weight on Read's experiment to determine that the inscription is not alphabetical. "Read ... says: 'It [the inscription] is precisely of such character as would be the result of an ordinary attempt to manufacture an inscription' and that any of the laborers could have made it. Read's well-structured and careful testing should have laid the question of the Grave Creek inscription to rest." (p. 86)

Williams' overall assessment of the stone, echoed in the title of his chapter, is, "Bah, humbug, Mr. Tomlinson." (p. 87)

Kelley

In a critical review of Williams' book, philologist David H. Kelley deals at length with Williams' treatment of the Grave Creek Stone. Kelly is the author of Deciphering the Mayan Script, a 1976 text that was instrumental in establishing the phonetic character of the Mayan glyphs. He considers as entirely appropriate Schoolcraft's much-ridiculed, if inconclusive, attempt to identify the alphabet of the inscription.

Kelley writes, "I have a hard time criticizing the view that the inscription is non-alphabetic, for that seems to me an obvious fantasy. I think that anyone who could not recognize that obvious fact should, ipso facto, disbar himself from any serious discussion of the problem. Williams praises an 'experiment' by Read to determine what geometric forms would be produced by a teacher, a schoolgirl, a druggist, and a college professor asked 'to write down twenty or more arbitrary symbols, not resembling any characters known to them and using only straight lines.' This rubbish is utterly irrelevant to the question of alphabets. If one can match an inscription to a specific alphabet or even to a closely related group of alphabets, it is alphabetic; otherwise, it is not. Inventing imaginary systems (by people familiar with alphabets) seems to have been a useful propaganda device, but such systems do nothing to support Read's conclusion that any of the labourers could have invented such an inscription. 'Bah, humbug, Mr. Read.'" (1995, p. 12)

Kelly goes on to discuss the similar Braxton Creek and Ohio County inscriptions mentioned by Fell, and concludes, "My major point, however, is not to argue that the inscriptions are, indeed, genuine, but rather that I do not find it fantastic to think that they may be. Williams's account makes the Grave Creek inscription look like obvious humbug, but he did not know many important facts which supported Tomlinson's account of the find, and he completely misinterpreted Schoolcraft's comparisons.... Williams's book .... should not be used as a bludgeon against looking at important but unusual data and trying to put such data in a genuinely appropriate archaeological context." (pp 13-14)

Smith

Most recently, Rev. C. Edward Smith, Jr. (1998) has made a detailed study of Fell's (1976b) translation of the Grave Creek inscription. Smith argues that Fell's interpretation of the text as Punic does not work at all, even taking his letter values as given. Personally, I couldn't tell Punic language from Arabic, but Smith cogently maintains that Fell doesn't seem to see the difference either, using as he does Wehr's Arabic dictionary to read Punic. Both are Semitic languages, but they belong to different branches of the family, and are evidently quite different in many pertinent respects.

Smith attempts no evaluation of Fell's interpretation of the letters themselves as Iberian script, nor of the authenticity of the stone. He concludes, "we simply do not know -- at this point in time and history -- what the Grave Creek Stone says -- only what it does not say."

Conclusion

Although it was amateurishly excavated, and years later there were numerous discrepancies about the details of its discovery, by all accounts it came out of the interior of the mound. Dr. Clemens documented the inscribed stone on the day of its discovery, even though Morton cut this account from the published version of Clemens' report. There is no particular reason to think the stone is fraudulent.

Although David Kelley regards the inscription as obviously alphabetic, there is still no solid confirmation of Barry Fell's identification of the script as Iberian. Fell's translation of the language as Punic has been sharply criticized by Edward Smith. Perhaps future study will establish a consensus about the alphabet, the language, and/or the message.

Davis, Emily C. "Printer's Knowledge of Dickens Solves Scientific Hoax," Science News Letter 17: 234-5, 1930.

Fell, Barry. America B.C. New York Times Books, 1976a.

Fell, Barry. "The Etymology of some Ancient American Inscriptions," Epigraphic Society Occasional Publications Vol. 3, No. 76 part 2, Sept. 1976b.

Hough, Philip R. "My Part in the Story of the Grave Creek Tablet," Tennessee Archaeologist Vol. 8 #2, Summer 1952, pp. 47-48.

Kelley, David H. "Epigraphy and Other Fantasies," The Review of Archaeology, vol. 15, #2 (April 19, 1995), pp. 8-14.

Smith, C. Edward, Jr. "What the Grave Creek Stone Does not Say: An Epigraphic and Philological Analysis." PO Box 96, Colerain, OH 43916. Revised Jan. 1998, June 2006.

Whittlesey, Charles. "Archaeological Frauds," Western Reserve and Northern Ohio Historical Society Tract No. 33, Nov. 1876, pp. 1-7.

Whittlesey, Charles. "The Grave Creek Inscribed Stone," Western Reserve and Northern Ohio Historical Society Tract No. 44, April 1879, pp. 65-68.

Williams, Stephen. Fantastic Archaeology: The Wild Side of North American Prehistory. University of Pennsylvania Press, 1991.