WARREN MOOREHEAD was a habitual diarist, and I was grateful. For when I read his diary entry for 16 December 1891, a crucial piece of evidence on a century-old archaeological fraud tumbled neatly into place. At last, the story of the Holly Oak pendant and its discoverer, Hilborne T. Cresson – archaeologist and Moorehead’s co-worker – was coming clear.
The Holly Oak pendant is a piece of whelk (a species of Busycon) shell, on which is engraved the figure of a mammoth. This animal was extinct in North America by 10 500 years ago, and if the engraving were made from life, it would be the only known example of North American ‘Palaeolithic’ art – something to show against the thousands of stunning Palaeolithic cave paintings and art mobilier of Europe.
First reported by Cresson in December 1889, and allegedly found in Delaware, this potential treasure had been rediscovered in this century in a specimen drawer at the Smithsonian Institution in Washington DC. In May 1976 the Holly Oak pendant adorned the cover of Science, with an article inside sug gesting the pendant might be as old as 40 000 years. An extraordinary find – but there were sceptics.
Advertisement
Among them was Bill Sturtevant, a curator of North American Ethnology at the Smithsonian. Bill has a keen eye and striking recall for artefact style and design. When he saw the pendant on the cover of Science, it reminded him instantly of the famous mammoth of La Madeleine, depicted on tusk, and discovered by Edouard Lartet in 1864 in the Dordogne region of France. When Bill read that the Holly Oak pendant came from Delaware, and not the Dordogne, he knew it must be a fraud.
I arrived at the Smithsonian in the summer of 1981 to begin a predoctoral research fellowship. Soon after our meeting, Bill learnt of my interest in the history of American archae ology and asked if, in my travels through 19th century archives, I would keep my eyes open for information on the pendant or its discoverer, Hilborne T. Cresson. I would, but I was not optimistic: Cresson was a peripheral figure in the archaeological community, the kind whose traces usually vanish after their time on stage.
Even so, I found a few obituaries. They revealed that Cresson was born Hilborne Jones, and assumed the Cresson name after marrying into that wealthy Philadelphia family. One notice referred to his ‘tragic death,’ but, oddly, none listed his age. I needed to know this, since he had been described as a ‘young man’ in 1889 when the existence of the pendant was first revealed, and yet he claimed to have made the discovery in 1864. How old could he have been in 1864? And why was the discovery only made public 25 years after the fact? Rather puzzling, the chronology of this story, and initially we did not appreciate the significance of the dates.
At the time, these meagre facts did little more than pique my curiosity, so I requested copies of a Philadelphia newspaper for September 1894, the month Cresson died, hoping at least for an obituary with a birth date. Two weeks later the microfilm arrived, with far more than I had bargained for. In bold type the headline for the 8 September issue of the Philadelphia Inquirer screamed the news: ‘DR. HILBORNE CRESSON TAKES HIS OWN LIFE’. I was stunned. As the paper indelicately put it, Cresson ‘blew his brains out in a park in New York City’. He had gone insane, his mind disordered, apparently the ‘result of scientific study,’ and ‘too close application of esoteric principles’.
The next week at the Philadelphia Public Library I learnt that Cresson’s suicide was carried on the front page of all four of Philadelphia’s major dailies. None knew his exact age at death (see Box overleaf). Most bizarre, among the items reported on his person at the time of his death was a note, in his handwriting, that he feared he was ‘suspected of counterfeiting and that Secret Service detectives were continually on his track’. We were getting somewhere.
Before he took his life, Cresson had been an archaeologist, working between 1887 and 1891 for F. W. Putnam at the Peabody Museum, Harvard University. Cresson got his start in Paris in the 1870s, where he studied ‘art and archaeology’ at the School of Fine Arts and Anthropology. Apparently, his talent for faithfully illustrating and recreating prehistoric art was considerable.
Cresson developed many interests in archaeology, and worked at several sites in North America. But for a brief time the object most dear to him, and of potentially the most significance to American archaeology, was the Holly Oak pendant. Here was a tidy solution, indeed, to the bitter dispute over the antiquity of humans in the New World.
Solving a Palaeolithic puzzle
From the 1870s American archaeologists had desperately sought the answer to whether the first Americans had arrived during the Pleistocene (more than 10 000 years ago) and were contemporaries of now-extinct mammals (such as mammoth), or whether their arrival predated Columbus by only a few millennia. The conflict exploded in the late 1880s, exposing the raw nerves of nationalism. Many wanted a prehistory just as deep and artistically rich as the Palaeolithic of Europe. Others wanted a new American archaeology, unshackled by expectations born of Old World Palaeolithic traditions.
One of the great champions of an American Palaeolithic was Putnam, Cresson’s employer. Putnam should have been delighted when, in December 1889, Cresson showed him the Holly Oak pendant. If authentic, the pendant clearly decided the battle in favour of a Pleistocene human presence in America. What better way to forge the ties between the prehistoric records of each continent, than to show that at the dawn of time the earliest Palaeolithic Americans created works of art to rival in age and beauty those of Palaeolithic Europe?
Yet, I searched Putnam’s papers in vain for the happy announcement that his cherished beliefs had been vindicated by this remarkable evidence. At Cresson’s request, Putnam showed the pendant at the meeting of the Boston Society of Natural History in February 1890. But, ever after, Putnam was steadfastly silent about the pendant.
Puzzled, I looked further into the contemporary literature. There was hardly any mention of the pendant in the thousands of published pages arguing for a deep human antiquity in America. Cresson himself never so much as alluded to it in print. (So much for claims that its discovery caused ‘great excitement’.) When none of the proponents of the American Palaeolithic used the Holly Oak pendant to further their cause, their silence was more damning than even the most vicious critics, such as archaeologist Henry Mercer who, busy peddling his own mammoth-engraving forgery, sneered that the Holly Oak pendant was a fraud and Cresson a liar.
The very day I was hunched over the microfilm readers in Philadelphia Public Library, Sturtevant was in Paris examining the original La Madeleine piece – and may be the only person since Cresson himself to have studied both original specimens. Bill’s detailed analysis of both pieces identified their stylistic parallels. But this was no surprise, since in the 1880s La Madeleine was one of the few examples of Palaeolithic art, and thus widely illustrated and an obvious source of copying.
Soon after seeing the pendant, Putnam prepared a photograph of it alongside a drawing of La Madeleine. Putnam may well have realised, even without knowing (as we now know) just how much variety there is in Palaeolithic depictions of mammoth, that the mammoths on Holly Oak and La Madeleine are uncannily alike, and in many idiosyncratic ways. We will never know whether Putnam believed, as we do, that Cresson simply backdated his discovery to early 1864 to have it occur before Lartet’s (another strange twist: Cresson claimed that his French tutor, allegedly involved in the pendant’s discovery, was a student of Lartet’s).
Nor will we know how Putnam reacted to Cresson’s explanation for the 25 years of silence between putative discovery and announcement. There had been similar lapses in two of Cresson’s unrelated but equally suspicious archaeological discoveries, and that bothered some of his peers. It meant Cresson’s finds could never be field checked. Perhaps Putnam just did not know what to think. After all, Cresson blamed the Holly Oak lapse on someone named ‘Mrs Spencer’, whom he identified variously as his mother, his sister, and a family friend.
In December of 1891 Putnam fired Cresson for stealing artefacts from one of his archaeological excavations. Sitting in Columbus, Ohio, reading the details of Cresson’s dismissal in Moorehead’s diary (Moore head was then Putnam’s field supervisor and uncovered the theft), I immediately remembered a talk I had a few months earlier with James B. Griffin, one of the doyens of American archaeology. Griffin thought how odd it was that the allegedly Pleistocene-age Holly Oak pendant, sans engraving, bore such a strong resemblance to shell pendants found in Fort Ancient period sites, which postdate AD 1000 (their precise age is still being derived). And in virtually the same flash I recalled that prior to publicly revealing the pendant in 1889, Cresson had worked for Putnam on Fort Ancient sites and museum collections. What better way to pass a forgery than to carve it on a genuine archaeological specimen that at least looked old?
Whether it was the disgrace of being summarily fired by his mentor that drove Cresson out of his mind, or the dismissal of his pendant, more through being ignored than publicly denounced, we shall never know. Regardless, Cresson’s suicide in 1894 would have been an end to this unfortunate story, but for one thing: any oral tradition of the Holly Oak pendant disappeared with the deaths of those involved. So it was, then, that the Holly Oak pendant was re-discovered in this century, with little knowledge of how it was received by Cresson’s peers. Unaware of its history, a Delaware geologist, John Kraft, and an archaeologist, Ronald Thomas, took to the field to document its potential age. By then, of course, archaeologists had established that the first Americans came here more than 10 000 years ago, but the earliest American art scene was still impoverished, and if truly ancient the Holly Oak pendant would stand as a stunning exception.
Kraft and Thomas carefully cored, sampled, and dated many of the geological strata in and around Holly Oak, Delaware. They showed there were deposits in the area that were at least 10 000 and perhaps as much as 40 000 years old. However, since they had not consulted archival documents to learn the circumstances of the find, the link of these ages to the age of the pendant was flimsy at best. Nonetheless, they suggested the pendant could be 10 000 or 40 000 years old, although made passing acknowledgment it might be a fraud. The editors of Science were clearly impressed, and displayed the Holly Oak pendant on the cover – the one that caught Sturtevant’s observant eye.
Our stylistic and historical analysis obviously led Sturtevant and me to a different conclusion: the pendant was a fraud. A long piece detailed our evidence (Anthropological Papers, no 72, University of Michigan). We also felt obliged to publish a brief comment in Science, and duly submitted one entitled ‘A mammoth fraud in Science’. Although we were rather pleased with the double pun, our editor informed Sturtevant the title must be changed to one that was ‘dull and not sensational’, then did just that. We were warned, too: Science was ‘investigating the legal ramifications of publishing (our letter) and . . . may request some additional changes if (their) lawyers so advise’. A charge of fraud is a serious matter. Sturtevant and I had not taken it lightly, and any journal understandably treads very carefully when such accusations are made. But alerting the attorneys on a case of possible fraud that was nearly a century old?
Our letter appeared in the 18 January 1985 issue of Science, followed by a response from Kraft and archaeologist Jay Custer. They were unmoved by our arguments. They were careful not to insist that the pendant (in their words a ‘potential national treasure’) was genuine, and admitted the possibility of fraud. But since the bulk of their piece attempted to demolish our arguments it was clear what they believed. Kraft’s position, it seemed, had not appreciably changed since he wrote that the pendant was ‘definite evidence of association of early American man with the woolly mammoth’.
Our historical, stylistic, and archaeological evidence, they thought, contained ‘nothing new or persuasive’. Perhaps so. But these were strong words, considering that until our work no one had consulted the extensive original archived documents on Cresson and the find, nor conducted the point-by-point comparison of Holly Oak and La Madeleine.
When we originally wrote our note, Sturtevant and I considered the possibility of directly dating the shell, but at the time the techniques for dating shell would have required the pendant’s complete destruction and yielded results of questionable validity. For that reason we were wary. But in the nearly two years between when our Science letter was submitted and when it was finally published, revolutionary developments were taking place outside the arena that would at last make it possible to date the shell directly.
Dating the undateable
The advent of radiocarbon dating using accelerator mass spectrometry (AMS) had now made it possible to count individual carbon-14 atoms in a sample, and thus obtain dates from tiny samples off objects – like the Holly Oak pendant and the Shroud of Turin that under conventional radiocarbon dating would have been essentially undateable. At the same time, researchers throughout the world were working together to calibrate radiocarbon dating, providing the potential for control over the vagaries of dating shell objects.
There was, of course, a risk in dating the Holly Oak pendant. What if we were wrong about this being a purloined Fort Ancient shell pendant, and Cresson (or someone else) engraved a well preserved Pleistocene age shell? But that prospect seemed remote and, confident in our historical and archaeological research, we predicted that the object, sans engraving, would date around AD 1000 (Fort Ancient times), not 10 000 or 40 000 years ago (Pleistocene times).
Even before our letter appeared in Science, we had applied to Smithsonian authorities for permission to date a few scraps of the shell (away from the engraving, of course) using the AMS technique. It would be another year, however, before tiny fragments of the pendant were off to the AMS laboratory in Zurich. During the long wait, Smithsonian archaeologist Bruce Smith, who arranged for the dating, organised a pool to guess the pendant’s precise age.
The resulting radiocarbon age of the pendant, when calibrated, came to AD 885, with a statistical range of AD 750 to AD 1000. Our prediction of its age was off by only a few hundred years, probably attributable to the margin of error that still exists in dating shell. Smith won the pool.
Kraft and Custer had speculated that ‘the last mastodon to die has not yet been found and may have died in the late Holocene’. Given the many hundreds of radiocarbon dates on mammoth and mastodon that have appeared in the past three decades, of which no reliable ones are younger than 10 000 BP (before present), that seems exceedingly improbable. Neither they, nor anyone else, believes that any of the now-extinct Pleistocene megafauna were alive a mere thousand years ago. There is no possibility that the Holly Oak mammoth could have been drawn from life.
Archaeology, not being a hard science, but a difficult one all the same, rarely yields the kind of unequivocal results obtained in the Holly Oak case. When it does, it is gratifying – regardless of whether one happens to be right or wrong. But let’s face it: most of us want to be right. Science is like that. After the date was published (American Antiquity, July 1988), many colleagues offered the opinion that they had known all along the pendant was a fake. Science is like that, too.
The story should end here, but doesn’t. A column in Science by Roger Lewin (2 December 1988) calling attention to our recently published date for the pendant triggered yet another response from Kraft and Custer (Science, 13 January 1989). They were highly critical of our discussion of Cresson’s personal history. Does that kind of information have a place in scientific discourse?
In our view, uncovering century-old fraud requires understanding not just of the object in question, but also the people and their times. Piltdown was a whopping big fraud, but what makes it fascinating is not just the specimen, but the fact that it was accepted for so long by so many people who really should have known better. To understand a scientific fraud – and perhaps even to uncover one – requires a firm grip on its context. That is good history. That is good science. And it certainly gave us the correct answer.
Cresson was terribly anxious to place his name on the rolls of American science. Yet the archaeological community was small, there were few salaried positions, and no obvious means to support the work necessary to acquire status – save, perhaps, for one: circumvent the process of hard work by making a discovery so fantastic that it will resolve the most bitter dispute facing American archaeology. The archaeological world will beat a path to your door.
The flaw in that strategy is that spectacular finds cannot be made on command. How desperate are you? If you are that desperate, you fake your own find. Science is not supposed to work like that, but sometimes blind ambition takes over, as we believe it did in Cresson’s case. So it was that a Fort Ancient shell pendant, in the hands of an artist schooled in French prehistory and with an illustration of Lartet’s La Madeleine before him, became for a moment the first American Palaeolithic art object. But only for a moment. Soon after Putnam fired him, Cresson began disowning his prior claims about the Pleistocene age of the deposits in which the shell had allegedly been found. He had no allegiance to the pendant. Just blind ambition.
Custer, contacted by a reporter after the latest squabble, admitted that the case was closing rapidly, but when asked about the pendant’s age replied, ‘Probably we’ll never know for sure.’ ‘And besides,’ he continued, ‘it’s sort of like, who cares?’ We may never know for certain – but only in the sense that dating a shell requires knowing where the shell came from, and on this point the trail is cold. We will probably have to be forever satisfied with a date of around AD 1000. However, that is good enough to seal the case. And, in truth, we all care about the Holly Oak pendant, and care very much. Knowing the pendant is a fraud has taught us not just about the late Pleistocene prehistory of America. It has also taught us a great deal about those – then and now – who practise the very difficult science of archaeology.
* * *
Mysterious origins of a putative fraudster
HOW old was Hilborne T. Cresson in 1864, when he claimed to have found the Holly Oak pendant? We never found out exactly, and although the fact that he probably hadn’t discovered the pendant that year made the question moot, we were still curious. None of the obituaries gave his date of birth, although a few of the newspapers estimated he was between 43 and 45 at death (in 1894).
To answer the question, I searched for Cresson’s descendants, but both of his sons apparently died without issue. I learnt, however, that there was a Cresson family plot in Oaks, Pennsylvania. Hoping for another lead, I telephoned Directory Assistance, but the only number I could get for Oaks was the Volunteer Fire Department. They didn’t answer.
An obituary of the elder son’s widow, the daughter of a sculptor, Daniel Chester French, opened a lead to the French family home, now the Chesterwood Museum. The curator there gave me the name of the keeper of the Cresson family bible, whose only comment on Hilborne was that he ‘was not a true Cresson’. He did not think Hilborne was – or should have been – buried in the Oaks family plot, nor did he know Hilborne’s year of birth. But he did give me the number of the small church in Oaks. The caretaker’s son answered the phone.
‘Yup,’ he said after a check of the cemetery records, ‘we’ve got him here’.
‘What are the dates on the entry,’ I asked.
‘Nothing in the records,’ he replied.
‘Let me check the headstone.’
I waited.
‘You’re not going to believe this,’ he told me when he got back on the line, ‘there’s no year of birth on the gravemarker.’
I drove four hours to Oaks that weekend and, sure enough, Hilborne T. Cresson’s grave is the only one in the yard without a birth date.
It was back to the US National Archives in Washington, DC to examine the decade Census Records – a daunting prospect and one I held out as a last resort. I started with the 1860 and 1870 censuses, but quickly abandoned them. Cresson was still a Jones then.
I had to look for Hilborne Cresson, which meant either the 1880 or 1890 census. I started with 1890, suspecting he was still in France in 1880. The 1890 census for Philadelphia, however, had been destroyed by fire. With little hope left, and beginning to sense a Tutankhamun-like curse, I turned to the 1880 census. There he was, living in Philadelphia with his Cresson in-laws. The story stayed peculiar to the end: Hilborne, who then had been a Cresson for only five years, is listed as the son and his wife as the daughter-in-law. His age was given as 32 as of 1 June 1880, making him 16 years old in 1864.
David Meltzer is in the Department of Anthropology at the Southern Methodist University in Dallas, Texas.