The Vinland map is claimed to be a 15th century mappa mundi with unique information about Norse exploration of America. It is very well known because of the publicity campaign which accompanied its revelation to the public as a "genuine" pre-Columbian map in 1965. In addition to showing Africa, Asia and Europe, the map depicts a landmass south-west of Greenland in the Atlantic labelled as Vinland; the map describes this region as having been visited by Europeans in the 11th century. Although it was presented to the world in 1965 with an accompanying scholarly book written by British Museum and Yale University librarians,[1] historians of geography and medieval document specialists began to suspect that it might be a fake as soon as photographs of it became available,[2] and chemical analyses have identified one of the major ink ingredients as a 20th century artificial pigment.[3][4] However, individual pieces of evidence continue to be challenged, most recently at a 2009 conference[update].[5]
Contents |
The Vinland Map first came to light in 1957 (three years before the discovery of the Norse site at L'Anse aux Meadows in 1960), bound in a slim volume with a short medieval text called the Hystoria Tartarorum (usually called in English the Tartar Relation), and was unsuccessfully offered to the British Museum by London book dealer Irving Davis on behalf of a Spanish-Italian dealer named Enzo Ferrajoli de Ry. Shortly afterwards, Ferrajoli sold the volume, for $3,500, to American dealer Laurence C. Witten II, who offered it to his alma mater, Yale University. It was initially treated with suspicion, partly because wormholes in the Map and the Relation did not match. In spring 1958, however, Witten's friend Thomas Marston, a Yale librarian, acquired from London book dealer Irving Davis a dilapidated medieval copy of volume 3 of Vincent of Beauvais's encyclopedic Speculum historiale ("Historical Mirror"), which turned out, amazingly, to be the missing link; the wormholes showing that it had formerly had the Map at its beginning and the Relation at its end. All traces of former ownership marks, except for a small part of a bright pink stamp which overlapped the writing on folio 223 of the Speculum, had been removed, perhaps to avoid tax liability for the former owner (although as historian Kirsten Seaver noted many years later, stamps on random book pages indicate institutional, not private ownership[2]).
Unable to afford the asking price, and concerned that, ostensibly because of the former private owner's tax concerns, Witten refused to reveal the provenance of the Map, Yale contacted another alumnus, Paul Mellon, who agreed to buy it (for a price later stated to be about $300,000) and donate it to the university if it could be authenticated. Recognizing its potential importance as the earliest map to show America, Mellon insisted that its existence be kept secret until a scholarly book had been written about it. Even the three authors of the book were chosen from among the small number of people who had seen the Map before Mellon bought it—two British Museum curators and Marston. Only one of them, Dr. Raleigh Ashlin Skelton, keeper of the Museum's map collection, had significant expertise relevant to the problems posed by the Map (his colleague George Painter, the first person to whom Davis had shown the map in 1957, was brought in for the transcription and translation of the Relation) and the secrecy almost completely ruled out consultation with specialists.[2] Witten did his best to help during this period, not only answering the authors' questions, but offering suggestions of his own.[6] After years of study, the proofs of the book, The Vinland Map and the Tartar Relation,[1] were ready by the end of 1964, and Mellon donated the Map to Yale. The book was published, and the Map revealed to the world, the day before Columbus Day, 1965. Many academic reviewers of the book took the opportunity to point out evidence that the Map was likely to be a fake, so a year later, a Vinland Map Conference was held at the Smithsonian Institution, during which further significant questions were asked, particularly of Witten, who gave very straightforward and helpful answers; but, the proceedings were not published for another five years.[7]
There were a number of questions about the actual content of the map. Witten had pointed out that it bore strong resemblances to a map made in the 1430s by Italian mariner Andrea Bianco, but others found some of the similarities and differences very strange—the Map cuts off Africa where Bianco's map has a page fold, but distorts shapes, and includes major revisions in the far east and west. The most surprising revision is that, unlike, for example, the famous Cantino World Map, the Vinland Map depicts Greenland as an island, remarkably close to the correct shape and orientation (while Norway, of which Greenland was just a colony, is wildly inaccurate) although contemporary Scandinavian accounts—including the work of Claudius Clavus in the 1420s—depict Greenland as a peninsula joined to northern Russia. For practical purposes, Arctic sea ice may have made this description true, and Greenland is not known to have been successfully circumnavigated until the 20th century. Skelton wondered also whether the revisions in the far east were meant to represent Japan—they seem to show not only Honshu, but also Hokkaido and Sakhalin, omitted even from Oriental maps in the 15th century.
In addition, the text uses a Latin form of Leifr Eiriksson's name ("Erissonius") more consistent with 17th-century norms and with transmission through a French or Italian source. The Latin captions include several usages of the ligature æ; this was almost unknown in later medieval times (a simple e was written instead), and although the ligature was revived by Italian humanist scholars in the early 15th century, it is found only in documents of deliberately classicising style produced by Italian scribes, and never in conjunction with a Gothic style of script such as is seen in the Map.
Another point calling the map's authenticity into question was raised at the 1966 Conference: that one caption referred to Bishop Eirik of Greenland "and neighboring regions" (in Latin, "regionumque finitimarum"), a title known previously from the work of religious scholar Luka Jelic (1863–1922). An essay by British researcher Peter Foote for the Saga Book of the Viking Society (vol. 11, part 1), published shortly after the conference, noted that German researcher Richard Hennig had spent years, before the Vinland Map was revealed, fruitlessly trying to track Jelic's phrase down in medieval texts. It seemed that either Jelic had seen the Vinland Map and promised not to reveal its existence (keeping the promise so rigidly that he never mentioned any of the other new historical information on the map), or that he had invented the phrase as a scholarly description, and the Vinland Map creator copied him. In practice, because Jelic's work had gone through three editions, Foote was able to demonstrate how the first edition (in French) had adopted the concept from the work of earlier researchers, listed by Jelic, then the later editions had adapted the anachronistic French scholarly phrase "évèque régionnaire des contrées américaines" into Latin.
Handwriting experts at the 1966 Conference tended to disagree with Witten's assessment that the Map captions had been written by the same person as the Speculum and Relation texts. This had also been a major reason why the British Museum had rejected the Map in 1957, the Keeper of Manuscripts having detected elements of handwriting style not developed until the nineteenth century.[2]
Complaints were made at the Conference that no scientist had been permitted to examine the Map and its companion documents in all the years of study since 1957. Skelton's scientific colleagues at the British Museum made a short preliminary examination in 1967 and found that:
a) despite its appearance to the naked eye, the ink was certainly not conventional iron-gall ink like its two companion manuscripts, and indeed was unlike any recipe they had ever seen (they spent several months after the initial testing, trying to find similar inks as far afield as Iceland);
b) the map outline appeared to consist of two superimposed lines, one black (but mostly vanished) which looked like graphite or soot, and one yellowish;
c) the whole of the Map parchment (again, unlike its companions) had been coated or soaked in an unknown substance—they were not allowed to take a large enough sample to analyze it;
d) they could not be certain that the two halves of the Map, held together by a binding strip glued on the back, had ever been a single sheet—unlike any other known medieval double-page map; looking at the map, it is clear that the artist knew exactly, to the nearest millimeter, where it was going to be folded, because several place-names start or finish right next to it while none are written straight across it, and the rivers of eastern Europe run parallel to it;
e) the rebinding of the Speculum volume without the Map and Relation had used plastic thread, only available since about 1950.[8]
In 1972, with new technology becoming available, Yale sent the map for chemical analysis by forensic specialist Walter McCrone whose team, using a variety of techniques, found that the yellowish lines contain anatase (titanium dioxide) in a rounded crystalline form manufactured for use in pale pigments since the 1920s, indicating that the ink was modern. They also confirmed that the ink contained only trace amounts of iron, and that the black line remnants were on top of the yellow, indicating that they were not the remains of a penciled guide-line, as the British Museum staff had speculated.[9]
A new investigation in the early 1980s, by a team under Dr. Thomas Cahill at the University of California, Davis, using Particle-Induced X-ray Emission (PIXE) found that only trace amounts (< 0.0062% by weight) of titanium appeared to be present in the ink, which should have been too little for some of McCrone's analyses to detect.[10] The Cahill team acknowledged, however, that titanium was the only element within their technique's measurement capability which was significantly more concentrated in the ink than on the bare parchment (other elements such as iron and zinc were found concentrated in some inked samples, but only a minority). One member of the team, Gregory Möller, also analyzed loose particles retrieved from the split down the middle of the Map by a different method, finding that most of them were rich in titanium (though a few black particles were rich in chromium and iron).[11] Because they were the first to apply PIXE to ink analysis, nobody at the time could explain the difference between the Cahill and McCrone figures. Attempting to reconcile the conflicting results, the Cahill team suggested that the high concentrations found by McCrone were due to a combination of contamination from modern dust, and poor sample selection (i.e. choosing contaminant particles like those in the split);[1] however, they also chose not to publish or publicise Möller's loose particle study.[12] The accumulation of large amounts of PIXE data from other laboratories around the world in the ensuing decades was sufficient by 2008 to show that the Cahill figures for all elements in the inks of the Map and its companion documents are at least a thousand times too small, so the discrepancy is due to their mistake.[3]
The McCrone team had also made mistakes, though none as fundamental as Cahill's. Revisiting his notes in 1987 to draft a detailed reply to the abbreviated public version of Cahill's report, Walter McCrone chose the wrong sample to illustrate a "typical" black ink particle, selecting one which had been found only loosely attached to the ink.[13] By focusing on this contamination, rich in chromium and iron, he gave Cahill the opportunity to re-emphasise his case in an essay for an expanded version of the 1965 official book, a few years later.
In 1991, McCrone visited Yale to take new microsamples from the map, partly to check his earlier results, and partly to apply new techniques. Photomicrographs taken at 1 micrometer intervals through the thickness of ink samples demonstrated that the manufactured anatase particles were not just sticking to the surface as Cahill's criticisms had implied, and Fourier transform spectroscopy identified the ink's binder as gelatin, probably made from animal skin.[14] In July 2002, using Raman spectroscopy, the presence of significant quantities of anatase in the map ink was confirmed by British researchers Katherine Brown and Robin Clark, and the remaining traces of black pigment in the ink were found to consist essentially of soot-type carbon.[4]
Various scientists have formed their own theories to explain how the "20th century manufactured" anatase in the Vinland Map ink could have got into genuine medieval ink. The first was chemist Jacqueline Olin, then a researcher with the Smithsonian Institution, who in the 1970s conducted experiments which produced anatase at an early stage of a medieval iron-gall ink production process. Examination of her anatase by a colleague, mineralogist Dr Kenneth Towe, showed that it was very different from the neat, rounded crystals found in the Vinland Map and modern pigments, and despite decades of further work, she was also never able to explain how the iron would have disappeared from the Vinland Map ink.[10][15][16] Towe himself, a clay specialist, briefly considered the possibility that the anatase could have come from clay, where it is present in trace amounts, but on checking McCrone's data found no significant traces of clay minerals. Shortly before the Raman analysis was published, historian Douglas McNaughton based a mistaken theory about the ink around McCrone's emphasis on the chromium-rich black particle, having obtained unpublished data on the similar particles in Möller's report.[17]
Radiocarbon dating, begun in 1995 by physicist Douglass Donahue and chemists Jacqueline Olin and Garman Harbottle, places the origin of the parchment somewhere between 1423 and 1445. The initial results were confusing because the unknown substance the British Museum had found across the whole Map, effectively ignored by later researchers who were concentrating on the ink, turned out to be trapping tiny traces of fallout from 1950s nuclear tests deep within the parchment. Although there is none of this 1950s substance on top of the ink, further tests, starting with a detailed chemical analysis, are needed to confirm whether the lines were drawn after it soaked into the parchment.[18]
In 2008, Harbottle's attempt to explain a possible medieval origin for the ink was published, but he was shown by Towe and others to have misunderstood the significance of the various analyses, rendering his theory meaningless.[3]
The expanded 30th anniversary edition of the 1965 official book, The Vinland Map and the Tartar Relation, was notable for its exclusion of most of the evidence against the Map's authenticity, concentrating instead on vindications by George Painter, and Thomas Cahill with colleague Bruce Kusko (in which they claimed specifically that they had not analyzed the loose particles they took from the map at the time of their PIXE research), but it did reprint a remarkable essay written in 1989 by the original book dealer Laurence Witten. He confessed that, when the McCrone evidence showed the map to be a forgery in 1974, he had been asked by Yale to reveal its provenance as a matter of urgency, and to discuss the possible return of Mr Mellon's money. He had replied that, contrary to all his previous detailed assertions, he had no idea where the Map came from, beyond Ferrajoli (who had been convicted of theft shortly after the sale, and died shortly after leaving prison). On the subject of the money, he had said he could not pay it all back because, again contrary to his very specific earlier statements, he had paid agreed shares of his profit to Ferrajoli and to another dealer who had introduced him. In the event, Mellon chose not to ask for the return of any money at all. The essay also revealed that Witten had, on Ferrajoli's recommendation, met with Irving Davis after buying the Map volume in 1957.[19]
Regardless of the controversy, the Map, which had been valued for insurance purposes at over $750,000 in the 1960s, was claimed in 1996 to be worth $25,000,000.[20]
In 2005 a team from the Royal Danish Academy of Fine Arts, led by Dr. René Larsen, studied the Map and its accompanying manuscripts to make recommendations on the best ways to preserve the centuries-old parchment.[21] Among other findings, this study confirmed that the two halves of the Map were entirely separate, though they might have been joined in the past. A few months earlier, Kirsten Seaver had suggested that a forger could have found two separate blank leaves in the original "Speculum Historiale" volume, from which the first few dozen pages are missing, and joined them together with the binding strip.[2] In this scenario, the mysterious chemical treatment of the parchment is intended to disguise slight differences of color and texture between the two halves, and the notch cut out of the bottom disguises a slight size difference (see bottom illustration). On the other hand, at the International Conference on the History of Cartography in July 2009, Larsen revealed that his team had continued their investigation after publishing their original report, and he told the press that "All the tests that we have done over the past five years — on the materials and other aspects — do not show any signs of forgery".[22] The formal report of his presentation[5] shows that his work ignores rather than contradicts earlier studies. For example, he experimented only with artificial wormholes, and did not follow up the observation made at the 1966 Conference, that live bookworms were a known tool of the fake antiquities trade. Similarly, he claimed that the anatase in the ink could have come from sand used to dry it (the hypothetical source of the sand being gneiss from the Binnenthal area of Switzerland) but his team had not examined the crystals microscopically, and Kenneth Towe responded that this was an essential test, given that crystal size and shape should clearly distinguish commercial anatase from anatase found in sand.[23]
Regardless whether the map is genuine, it is known that Greenland was settled by Vikings around 970, a settlement which lasted until the 15th century. In regard to the Americas, the archaeological finds in L'Anse aux Meadows in present-day northern Newfoundland, Canada—to which the investigators had been led partly by the 16th-century Skálholt Map[24]—show there was a Viking settlement there (around 1000) which, while unsuccessful and short-lived, predates by five centuries Christopher Columbus's voyage to the Caribbean islands in 1492 and John Cabot's voyage to North America in 1497.
As controversy has swirled around the map almost since its acquisition, authorities at Yale University have chosen not to comment on the authenticity of the parchment document, other than to say they watch the debate with unusual interest. "We regard ourselves as the custodians of an extremely interesting and controversial document," said Yale librarian Alice Prochaska in 2002, "and we watch the scholarly work on it with great interest."[25]
