054
Let me see if I have this straight. Some 19 centuries ago there was a group of Jews, citizens of one of the Judean port cities like Caesarea or Joppa, who fled their homes to escape the violence and confusion of the First Jewish Revolt against Rome and, reluctantly forsaking their native land, embarked on a life at sea. They probably turned to piracy for a while, preying on commercial vessels in the shipping lanes of the eastern Mediterranean, like the refugees from Joppa whose shipwreck is reported by Josephus in The Jewish War.1 Eventually, though, they left everything familiar behind and sailed west, possibly in search of more vulnerable quarry farther from the patrols of the imperial fleet. In all likelihood, they plied their trade for a while in the western Mediterranean, though we have no way of knowing exactly where. Perhaps they operated off the Iberian shore for a while; perhaps they spent some time (in what was soon to become the best piratical tradition) loitering along the Barbary Coast. In any case, we can be fairly certain that they were headquartered far enough west to be able to make an occasional foray beyond the Pillars of Hercules. On one such occasion, we must suppose, they were surprised by a sudden squall and swept out into the open Atlantic.
It is not difficult to imagine the state of mind of this little band of Judean expatriates when they found themselves on the high seas in foul weather and with no land in sight. They must have been skilled enough mariners to know better than to head back toward shallow water before the storm abated, but they must also have known, if only from seafarers’ tales, of the heavy seas and relentless gales of the North Atlantic. So, despite their confused and frightened condition, they had the presence of mind to try to sail south. As the weather cleared, they received the welcome help of the northerlies that prevail off the coast of Spain and North Africa and, running ahead of these favorable breezes, found themselves fairly quickly in the vicinity of the Canary Islands.
Did they rest in the Canaries? Did they encounter another storm? Did they decide that, having come this far, they might as well go on? We have no way of knowing, but under whatever circumstances they headed west and (assuming the season was right) sped along before the trade winds that blew steadily from the east. Though their course anticipated the one Columbus set more than a millennium later, they somehow missed, or at least passed, his point of landfall at San Salvador as well as the larger and smaller cays of the Bahamas and shot through the Straits of Florida in the face of the sultry current that flows out of the Gulf of Mexico. They must have felt very much at home after that, sailing in relatively calm water and hugging this new coastline as if it were that of their native Mediterranean.
In the course of time, their explorations took them into the mouth of the Mississippi River. Sailing north, they may have thought of the Nile as they sailed past places that would later be called Alexandria, Memphis and Cairo. Some 600 miles inland, near the future Cairo, Illinois, they turned east onto the Ohio River, then south onto the Tennessee. Though its navigability gave out somewhere in northern Alabama, they continued to follow this river system up into the foothills of the Great Smoky Mountains, until they reached what would someday be the Little Tennessee River in Loudon County, Tennessee. From here they traveled no more.
One day, one of the people who had made the voyage picked up a flat brown stone and scratched several letters onto its surface. He doesn’t seem to have retained very much of the writing of his native Judea, which understandably had become rather remote in his memory, but what he did remember was enough to impress the locals, Woodland-period (1000 B.C.–700 or 800 A.D.) mound builders, who came to cherish that little stone enough to include it among the grave goods in a nine-person burial at the place where Bat Creek empties into the Little Tennessee River.
I like this story. I hope that it or one of the countless variants of it that can be imagined—is true. But it seems unlikely. Very unlikely. Let’s consider why.
The archaeological phenomenon we are confronted with is this: A paleo-Hebrew inscription has been found in a controlled archaeological context in a Woodland burial mound in Loudon County, Tennessee. This means that, if the archaeological context is truly controlled, an artifact inscribed with a type of writing native to Hasmonean and Herodian Judea somehow found its way into the hands of contemporary people living in the New World. As the fictitious scenario sketched above shows, this is not impossible; but any reasonable person will have to admit that it is so extremely improbable that it can be maintained only on the basis of very special pleading. Well, it has been maintained—and by no lesser authority than the eminent Semitist Cyrus Gordon, among others. Why?
There is one (and only one) thing about the Bat Creek stone that causes a Semitist to take a second look at it. This is the close superficial resemblance of a sequence of five of the signs (letters 3 through 7; see the drawing in the sidebar “Deciphering the Bat Creek Inscription,” in this issue) to the signs representing the sequence lyhwd in certain periods of ancient Hebrew writing. Though, as explained below, this sequence by itself cannot be read as Hebrew from the time of the First Jewish Revolt (66–70 A.D.), it can be interpreted as a part of certain expressions that read quite straightforwardly, namely, lyhwdh, “to Judah,” and lyhwdym, “to (the) Judeans/Jews.” Moreover, five letters in a row seem a bit too many to attribute to an accident involving coincidental letter forms, especially when these letters produce a linguistically intelligible sequence. So, these letters make the Bat Creek stone interesting.
There are, however, paleographical difficulties with the forms of the five letters, some of which do not correspond to their proposed paleo-Hebrew prototypes closely enough to be considered authentic (he [letter 5], waw [letter 6], dalet [letter 7]). Considerations of this kind have already been raised by Frank Cross, whose observations McCulloch has attempted to rebut in detail.2 So, for example, when Cross objects to the form of the alleged Bat Creek he (letter 5) as “impossible in the period 100 B.C.–A.D. 100,” McCulloch responds by calling this a “clearcut error,” citing an example of what he considers an “essentially identical” paleo-Hebrew he from Mark McLean’s 055doctoral dissertation (see chart below), which Cross himself directed. In fact, however, although this he may look similar to an untrained eye, it is quite unlike the Bat Creek sign, most especially because it has a clear vertical stem extending below the bottom horizontal, as is always the case with the paleo-Hebrew he. There does not seem much point in reproducing here the other details of the exchange between Cross and McCulloch, except to say that after looking it over in detail, it strikes me that Cross’s analysis is reasonable and convincing. (This is not, I admit, a bold conclusion on my part, since Cross is, after all, the acknowledged world authority on the subject at hand.)
But even if there were no paleographical problems with the sequence lyhwd, the meaning of the sequence would present difficulties. The traces of the sign (letter 8) that follows this sequence cannot be interpreted as a paleo-Hebrew he under any circumstances, and this rules out the reading lyhwdh, “to Judah.” Gordon’s suggestion that it be completed with a mem, giving lyhwdm, “to (the) Judeans/Jews,” can be accepted only on the unlikely assumption that the writer omitted a yod (y) while intending to write lyhwdym. McCulloch’s lyhwd, “to Yehud/Judea,” is ruled out by other considerations. Yehud was a name used in the Persian period (538–332 B.C.) for the district of the Persian empire that corresponded to Judea in the Hellenistic and Roman periods, and though it appears commonly as yhwd on coins and seals of the late Persian period (i.e., the fourth century), it would be out of place on an artifact from the time of the First Jewish Revolt. McCulloch’s appeal to a personal name in a paleo-Hebrew tomb inscription (the Abba inscription) is beside the point, since it is not simply a question of orthographic convention, as he seems to understand it, but of the currency of the name itself. It would be as if a contemporary citizen of New York should refer to his home as New Amsterdam.
As McCulloch himself admits, the other signs on the Bat Creek stone are poor candidates for comparison with paleo-Hebrew, despite his and Gordon’s various attempts to explain them in that way. It seems clear, instead, that we have an object which is not an authentic paleo-Hebrew inscription, but which clearly imitates one in certain features. How is this to be explained?
It is probably not a case of the coincidental similarity of random scratches to ancient letters, since, as noted above, the similarity extends to an intelligible sequence of five letters—too much for coincidence. It seems probable that we are dealing here not with a coincidental similarity but with a fraud.
Having said this, however, I am obliged to say something about the radiocarbon determination of 427 A.D. (with a 95 percent confidence interval of 32 to 769 A.D.) obtained by McCulloch for wood fragments found in the tomb in association with the inscribed stone. When this test was made in 1988, its results were sufficient to raise the Bat Creek stone out of academic obscurity for fresh consideration. The results of the test on the wood do not, however, establish the antiquity of the stone. There has never been any question that much of the material in the Bat Creek mounds derives from an early period, so the wood may well have been contaminated from other materials in the wet environment of the mound. But even if we assume, for the sake of argument, that the wooden fragments are as old as the carbon-14 test indicates, the relevance of their date to that of the stone depends entirely on the integrity of their association with it. And if, as I’ve already suggested, this is a case of fraud, that integrity can hardly be assumed.
Who may have perpetrated such a fraud I don’t know, but people closer to the scene in Loudon County have made some interesting suggestions. Most revolve around the figure of John W. Emmert, the man who found the Bat Creek stone in 1889 while working for the Smithsonian Bureau of Ethnology. According to Lowell Kirk, a historian at nearby Hiwassee College,3 the whole affair was a spinoff of the local politics of East Tennessee. Professor Kirk suspects that the miscreant was a Republican tombstone maker-and displaced Yankee to boot!-by the name of Luther Blackman, who carved and planted the stone in an attempt to discredit Emmert, who was an East Tennessee Democrat and political employee. More often, however, Emmert himself is suspected of being the culprit.4
Between the years 1883 and 1889, Emmert was given a large number of assignments by the Smithsonian. Although he persistently sought to secure a permanent, full-time post, he had so far failed to do so, perhaps in part because of an excessive fondness he is said to have had for the good corn whiskey for which his home state is renowned.5 The reliability of some of the conclusions reached in his earlier excavations was doubted by his employers, and his conduct in the Bat Creek affair has been questioned accordingly. His work on the Bat Creek excavation was funded for the specific purpose of testing the hypothesis of Cyrus Thomas, Emmert’s immediate superior in Washington, that much of the moundbuilding in the eastern United States was done by ancestral Cherokees. Evidently, Emmert got himself a book containing drawings of ancient inscriptions, used it to forge the writing on the Bat Creek stone, pretended to find it sealed in one of the mounds and sent it to Thomas, who proclaimed it an early example of Cherokee syllabic, a palpable confirmation of his own hypothesis about eastern moundbuilding.
If Emmert’s purpose was to achieve this result, thereby ingratiating himself to his boss,6 his choice of a model was ironically inept. When the counterfeited paleo-Hebrew signs were noticed, they were used as evidence of early contacts with the Old World, lending support to scholars who looked in that direction instead of to early Cherokee inspiration for an explanation of the development of American moundbuilding. It therefore seems more likely that Emmert chose an Old World model with the calculated purpose of embarrassing Thomas and others at the Smithsonian who had denied him permanent employment for so many years.
In any case, the Bat Creek stone has no place in the inventory of Hebrew inscriptions from the time of the First Jewish Revolt against Rome. Its history belongs to the melodrama of American archaeology in the late 19th century, rather than to the mystery of trans-Atlantic crossings in remote antiquity.
