Out of Eden: The Peopling of the World
Page 28
The strength, and at the same time the chief weakness, of the Clovis-first theory was that the earliest Clovis point had to be just after the opening of the ice corridor, and the ice corridor had to be just before the earliest Clovis point. Like a house of cards, nudge a key structural element and it will fall down. The key element was the insistence on the dates limiting the earliest entry to the Americas to after the opening of the ice corridor. If the occupation of North or South America could be pushed back just a few thousand years to a time when the corridor would have been closed, the theory would fall. In that case, the first entry would have to have been much earlier – before the Last Glacial Maximum (LGM), and probably before 22,000 years ago.
Pretenders, heretics, or scientists?
Seen in this light, new pretenders bearing dates more than 1,500 years before Clovis would in effect be suggesting entry to the Americas before the ice age rather than after, a radically new theory. Although it would be marginally less persuasive, any evidence of occupation in South America, even slightly before Clovis and with different adaptive technology, would also make it unlikely that Clovis points identified the entry of the first Americans.
There have been plenty of new pretenders over the past few decades, whether we are talking about new archaeological sites with pre-Clovis evidence, or their advocates. In one review from 1990 I counted eighteen contender sites. This is a conservative estimate, perhaps half, of the total challenges mounted over the past two decades. Most of these pretenders have been routed by defenders of the Clovis-first orthodoxy in skirmishes characterized by attacks on scientific method, context, and observation. Only a few have survived the ‘critical’ broadsides fired by defenders of the orthodoxy, but they are still embattled. The most persistent and serious pretenders are the sites of Monte Verde in northern Chile and Meadowcroft Rockshelter in south-western Pennsylvania (see Figure 7.1). New recruits have joined their ranks, namely Cactus Hill in Virginia and Topper/Big Pine in South Carolina.6
Figure 7.1 Theories of entry to America. At the LGM, access to the Americas was blocked by ice. Two window periods existed, one just before and one just after the LGM, when the Beringian land bridge was open and access was possible through the ice corridor or along the west coast. Also shown are other theories and sites mentioned in text. For clarity, present-day coastline shown.
The history of these claims and refutations is revealing of academe. The weapons of argument used by the defenders certainly appeal to scientific method, careful observation, avoidance of known sources of error, balanced logic, and reason; but the tactics of appeal are clearly biased, selective, petty, personal, and confrontational. In other words, the method of defensive attack is ‘sling enough mud in the form of possible errors or what if’s, and the good evidence all becomes tarred with the bad and can be dismissed’. This is reminiscent of a stock courtroom drama, where the witness is discredited by the clever, aggressive lawyer. That is, however, an adversarial, not a scientific approach. Furthermore, unlike the judge, archaeologists are not required to come to closure of a case or theory. They can and should remain open to different interpretations of the evidence. The ‘truth’ – whatever it is – is not on trial. Clovis-first and its defenders, however, may well be.
There is no particular a priori reason to think that America was first colonized after the last ice age rather than before, since sea-locked Australia, New Guinea, and even the Bismarck Archipelago and the North Solomon Islands were all colonized well before the LGM. On the other hand, there is every reason to assume that evidence is generally clearer the more recently it was set down, and that the global effects of the ice age destroyed much good evidence in North America. The fact, therefore, that there are a lot of Clovis points lying around from a few thousand years after the LGM, in good context, as opposed to the less well-provenanced evidence from before the LGM, does not prove that Clovis was first. It merely disproves the preceding orthodoxy of the nineteenth century, that America was not colonized until 10,000 years ago. We should expect the preglacial evidence (if any) that questions Clovis-first to be weaker than Clovis itself.
Monte Verde
We can illustrate some of these points with the history of the Monte Verde controversy. Tom Dillehay of the University of Kentucky has been involved in the excavation of Monte Verde site in southern Chile since 1977, and he and his colleagues have amassed a considerable arsenal to use against the Clovis-first citadel. While there is some evidence for a very early occupation of Monte Verde, with split pebble tools dating to around 33,000 years ago, the ‘best’ evidence is dated much more recently. The site is a peat bog and has preserved a number of organic and other remains suggestive of human occupation: a footprint; wooden artefacts; structures that may be parts of huts; hearths; the remains of palaeo-llamas and mastodons, including cut bone; and seeds, nuts, fruits, berries, and tubers. Radiocarbon dating of the organic remains has yielded dates from 11,790 to 13,565 (average 12,500) years ago. Simple stone tools such as flakes and cobbles were also found.7
Monte Verde lies 12,000 km (7,500 miles) south of the Alaskan ice corridor and, with an age of around 12,500 years, appears to antedate any Clovis site by a millennium. This raises the logical question of how anyone could have passed through the ice corridor after 13,000 years ago and still had time to travel so far south and then change culture.8 This looked like powerful ammunition to use against the Clovis-first orthodoxy. Predictably, the Clovis-first camp fought back. To break the resulting deadlock, an independent set of referees was brought in.
In 1997 an invited group of the foremost Palaeo-Indian specialists, including sceptics such as Vance Haynes, gathered in Kentucky to hear Dillehay’s presentation, and then visited Chile for a site inspection and further presentations. Each member of the group was handed a detailed site report published by the Smithsonian Institution. The group’s consensus report, published later that year in the academic journal American Antiquity, concluded that the site was an archaeological site and that the dates of occupation were around 12,500 years ago. A further article in American Antiquity by dating specialists, including Vance Haynes, confirmed the dates, freeing them from any suspicion of contamination with older carbon sources.9
Then something happened which revealed the dispute in its true colours. This was now an academic turf war, with no place for the open-mindedness, objectivity, and reason expected of scientific peer review. Just when the Dillehay camp thought that the matter was closed, the wall of scepticism broken, and Monte Verde finally accepted by the establishment, a private archaeological consultant, Stuart Fiedel, published a long and stinging critique of the entire site report. Media drama intensified. Unconventionally, Fiedel chose to make his attack not in a peer-refereed academic journal but in a popular magazine, Discovering Archaeology.10 The journal also ran responses from Dillehay’s group and supporters, as well as further sceptical comment and what amounted to a partial retraction by Haynes. This time the reaction came not just from the Dillehay supporters but from the establishment, in particular from Archaeology, the high-profile, high-circulation organ of the Archaeological Institute of America. The strength of the reaction is evident from the following extracts from both publications. First, from Discovering Archaeology:
If the case for the new idea or theory is deemed convincing, then the old paradigm quietly dies and the particular scientific field is better for it. In point of fact, however, and particularly in archaeology, it seems this admittedly idealized sequence of events rarely comes to pass. Instead, the whims of personality and pride are frequently injected into the process, acrimony and ad hominem attacks set the tone of the debate, and objectivity is submerged in personal polemics. For these reasons, paradigms, especially old ones, die harder than Bruce Willis. This is clearly the case with the ‘Clovis-first’ or ‘Clovis primacy’ model, which is now more than 50 years old. (James Adovasio, ‘Paradigm-death and gunfights’11)
[Fiedel] blankets the whole in a patina of almost-conspiratorial mis
trust and accusations, layered with all too-frequent snide remarks (e.g., ‘Dillehay’s Hamlet-like agonizing’). In that, he does his critique no favors. More to the point, the most useful, productive, and constructive procedure (and certainly the most open and fair minded) would have been to first send Dillehay a draft of the criticism for comment and clarification of what were the trivial/editorial problems, and then once the minor problems were dispersed with and the major issues clarified, submit the piece to a rigorously peer-reviewed academic journal. Fiedel did not do so. (David Meltzer, ‘On Monte Verde’12)
And from Archaeology:
Fiedel’s review is clearly biased and negative in tone. He ignores material that does not support his critical thesis and takes the more negative or improbable of alternative views of each case that he discusses. (Michael Collins, ‘The site of Monte Verde’13)
I’m irked at the carping tone of Fiedel’s commentary, and the ferreting out of meaningless conflict in interpretation over two decades of reporting on Monte Verde. Fiedel cops an attitude which, in my opinion, is entirely inappropriate. For my money, the Monte Verde research team should be celebrated, rather than henpecked, for their willingness to publish their findings in great detail and to share their misgivings about their own data . . . I think its [sic] a cheap shot to dredge up preliminary assessments and press reports to attack the Monte Verde project. I still think its [sic] a good thing to change your mind (so long as you’re honest about it). I question Fiedel’s decision to rush his manuscript into print without first passing it by Dillehay and his colleagues for comment. Picking up on my earlier theme of self-criticism, its clear that the mishmash of Fiedel’s shotgun criticism could have been winnowed down through frank person-to-person communication with the Monte Verde principals. Once this interchange had occurred, we could have been presented with a concise summary of the real issues, unclouded by the chaff and attitude . . . Ambushing the Monte Verde team in this way inevitably raises questions over Fiedel’s motivations. Was the critique primarily concerned with clarifying the pre-Clovis possibilities at Monte Verde? Or was this just another carefully timed headline-grabber? Handled the way it was, who knows? (David Thomas, ‘One archaeologist’s perspective on the Monte Verde controversy’14)
Other reviews of Fiedel’s critique were approving. But the furore it caused left the big question mark hanging over Monte Verde.
For the interested lay person, or even an archaeologist unfamiliar with the sites, trying to assess what was going on here is rather like having to choose a doctor when all the doctors are accusing one another of professional misconduct. One clear conclusion from that hypothetical situation would be to distrust doctors’ opinions of one another. Perhaps the shrillness of the defenders of the Clovis-first orthodoxy is a measure of how much they perceive Monte Verde as a threat. Their response also serves as a baseline for assessing objections to other sites that are claimed to be pre-Clovis. If the objections come from well-known defenders of the Clovis-first orthodoxy, distrust them. If the objections (or confirmations) come from neutral experts or from the whole profession, take them seriously.
Meadowcroft
Another veteran high-profile site under siege is Meadowcroft Rockshelter. For thirty years, Pennsylvania archaeologist James Adovasio has led work on this site (see Plate 24). He and his colleagues have dug through eleven floor layers, unearthing 20,000 stone flakes and objects and a huge quantity of animal and plant remains. Fifty-two radiocarbon dates have been published for the Meadowcroft finds, the oldest at the bottom in sterile clay 31,000 years ago and the youngest at the top, just 1,000 years old. Dates unambiguously associated with Palaeoindian occupation go back to 16,225 years ago, while dates in excess of 19,000 years have been claimed for the deepest occupation layer.15
As soon as these old dates were published they drew a storm of protest. No prizes for who has been the fiercest critic: Vance Haynes. Years of criticism focused on details of stratigraphy, documentation, dating anomalies, and possible contamination by a coal seam a kilometre away. Adovasio is reported as regarding such criticism as pathological scepticism. Thousands of pages have been written answering specific questions. The contamination issue was buried by an independent geomorphologist in 1999, but Haynes still wants to get carbon dates on a few remaining items, a nutshell and some seeds. Adovasio has had enough, however. He is reported to have informed Haynes three years ago at the Monte Verde meeting that, ‘I will never run another date you have asked me for, because since 1974, we’ve addressed every criticism anyone has raised. I have spent half my life on this.’16
If professional sceptics such as Haynes and Fiedel are right, then they are to be congratulated for maintaining their integrity in the face of a massive synthesis of false evidence. If they are wrong and/or biased in their approach, on the other hand, then they would have successfully and vaingloriously held up the progress of American archaeology for three decades and artificially prolonged the life of the obsolete Clovis-first orthodoxy to around seventy years. With the ever-accelerating pace of scientific discovery, that would be an extraordinary achievement. Even Hrdlicka did not achieve that length of filibuster at the beginning of the last century.
Two other, more recently discovered North American pre-Clovis sites now vie for immediate attention – Cactus Hill and Topper. Cactus Hill, on the East Coast near Richmond, Virginia, now has a full site report, written by two competing private archaeological teams, waiting for the sceptics to tear into it.17 Joseph and Lynn McAvoy of the Virginia Department of Historic Resources run one team, while Michael Johnson of the Archaeological Society of Virginia runs the other. Cactus Hill is an ancient sand dune, so-called after the prickly pears that cover it in the summer.
The original find at Cactus Hill was made, as often happens, by a perceptive farmer; he noticed a stone point in a pile of sand dumped some way from the hill and traced it back to its source. Digging through the layers of time, the teams have uncovered stone points of progressively greater antiquity including some of the fluted Clovis type. At the lowest level they found flaked tools, a scraper, a quartzite core, and some small blades. Radiocarbon dates from this bottom layer are in the 15,000–16,000-year age range – earlier than the opening of the ice corridor. In one of the deep layers, unusual stone points suggested an antecedent to the full Clovis style. Together with the other tools, these finds constitute even more evidence for a pre-Clovis culture. Needless to say, Haynes and Fiedel have been here too, to doubt the age of the artefacts. They have not, of course, suggested that they are older, rather that they are younger.18
The other new site is Topper, in South Carolina, named after David Topper, the forester who discovered it. The site research is directed by Al Goodyear, a state archaeologist with the University of Carolina. Five years ago Goodyear and his team were forced off another site by floods and re-excavated Topper instead. Only this time they dug deep, deeper than the Clovis level. Formerly an uncontroversial card-carrying Clovis-first archaeologist, Goodyear was converted by the shock of what he found. Below the Clovis level were small blades of chert, chiselled burins, a scraper, and micro-blades. The technology was more reminiscent of the Upper Palaeolithic in Siberia than anything previously found in the American south-east. Luminescence dating puts the age of the artefacts at 13,000 years.19
Have the sceptics chimed in again? Of course. This time their beef is not with the dates but with the tools. Were they made by humans? Another objection: most Clovis sites do not have pre-Clovis tools beneath them. According to the journal Science, Vance Haynes finds it hard to accept that this is just ‘a coincidence’. Such an argument seems rather like rejecting the Roman occupation of England because not every house in England has Roman remains beneath it. ‘I’ve been looking at this for 40 years’, says Haynes. Hmm.20
Two more recent sites with pre-Clovis dates are playing out their cycles of claim and criticism: Schaefer and Hebior, in south-eastern Wisconsin. Both sites have been radiocarbon dated to around 12,500 year
s ago. At one of them, Hebior, near Kenosha, flakes, a chopper, and two flaked stone bifaces have been found among butchered mammoth bones.21
With archaeologists such as Dillehay, Adovasio, the McAvoys, Johnson, Goodyear, and their colleagues going to extreme lengths to document their evidence, warts and all, and the persistent ‘what if’ responses from the critics, there has been a sea change in discussions on the pre-Clovis issue. The critics are at last being viewed less as the careful counsel of the establishment, and more as those who ‘doth protest too much’.
The Clovis-first mindset has, at last, been weakened. There is now a rash of recycled, new, or alternative explanations of the early colonization of the New World. The debate is moving from asking whether there was a pre-Clovis movement into and throughout the Americas, to which, and how many, exotic routes were taken. These ‘big arrows into America’,22 which, if all valid, would have to imply multiple entries from elsewhere, including a west coastal land route, a west coastal sea route, a North Atlantic route from Europe, a South Pacific route from Australia, and a South American re-entry after the ice age (see Figure 7.1). There is, however, less secure archaeological evidence for any of these scenarios than for Monte Verde and Meadowcroft.