This Argument Has A Few Holes In It. 80-odd-Thousand-Year-Old Shell Beads From La Grotte des Pigeons, Morocco and Blombos Cave, South Africa

Grotte des Pigeons in eastern Morocco, near the village of Taforalt. Photo by Abdellah Elbadaoui (Panoramio/GoogleEarth).

As I continue to wend my way through the literature I've missed during my absence from palaeoanthropology, I'm forever discovering scholarship that makes me think I'm on the other side of Alice's looking glass. Yesterday, while nosing around in search of background knowledge for the Subversive Archaeologist, I found another beaut that I can't ignore. And today I found it a companion. The two articles concern recovery of perforated marine shells from about 80 ka at the site known as Grotte des Pigeons [pictured above], near Taforalt in Morocco and at Blombos Cave, in South Africa. Of greatest interest to me is the way the authors attempt to rule out natural processes in pursuit of an explanation for their presence in the caves.

I hope to show why their efforts are insufficient, and why we should treat with caution these claims for modern human behaviour at such early dates. Remember Rule #1? Rule out natural site formation processes before invoking human or hominid activity. Its corollary is this: to rule out natural processes you need to keep in mind that different processes can result in similar archaeological 'signatures.'

For example. Remember Raymond Dart and the "osteo-donto-kerartic culture" of the australopithecines? Dart had observed the behavior of members of traditional societies in southern Africa. He noted that when attempting to extract marrow from the long bones of their prey, they employed what Dart referred to as "the crack and twist" technique. The marrow seeker would strike the shaft hard enough to crack it. Then, grasping one end in each hand, the hunter would twist the ends in opposite directions until the shaft broke open. On the basis of these observations Dart proposed that the presence of similarly broken fossil bone meant that the Australopithecines had been hunting, then cracking and twisting the long bones to extract the marrow. It was many years before someone pointed out that the same effects can result from carnivore modification, and that it was much more likely that the presence of bones broken in that way was the result of carnivore behavior, rather than that of small, bipedal apes. Dart had premised his explanation on the only analogous process of which he was aware at the time. Such shortcomings permeate the archaeological canon. They remind us that we must always consider the possibility of equifinality---the idea that more than one process can result in a given set of observations. Moreover, to determine whether your observations are humanly or naturally produced, you need to involve as broad a range of present-day analogies as is possible. In the case of Grotte des Pigeons and Blombos Cave, the archaeologists have posited evidence that is sufficient to explain what they recovered, but their evidence is by no means the only possible explanation, nor necessarily the most parsimonious. At best their evidence is equivocal [a accusation that I've used over and over again in my work].

Enough preaching. On to the sea shells.

At La Grotte des Pigeons and at Blombos Cave excavators recovered small marine invertebrate remains that were altered from their natural state. In all, 13 were recovered from Pigeon Cave, and 41 from Blombos. The shells are perforated, and some bear a polish, which, the authors propose, you might expect if these objects had been strung like beads on a necklace. In addition, some very small patches of iron oxide were found on a few, which suggests to the authors that the putative shell beads had been artificially coloured with ochre. In the illustrations below you see the entirety of the collection from La Grotte des Pigeons. [I'm impressed by the presentation---for each artifact all six aspects are illustrated. Compare this to the artful display of the Blombos shells, further down, which is arranged so as to illustrate the path through which some twine might have passed.]

When viewing these images, keep in mind that the shells themselves measure about one centimeter across and slightly more in the longest dimension. Thus, the perforations are on the order of 500 microns in diameter. Note also that a great many of the perforations are irregular, and look more like they've been broken open rather than having been drilled.

From Bouzouggar, A., N. Barton, et al. (2007). "82,000-year-old shell beads from North Africa and implications for the origins of modern human behavior." Proceedings of the National Academy of Sciences 104(24): 9964-9969. Specimen 14 is a modern-day individual.

The photo montage below seeks to illustrate the areas of 'unnatural' polish and other markings on several of the Pigeon Cave shells. The authors claim that these are likely to be the result of having been strung like a strand of ... well, beads. I'd have to say that these aren't patterned alterations.

From Bouzouggar, A., N. Barton, et al. (2007). "82,000-year-old shell beads from North Africa and implications for the origins of modern human behavior." Proceedings of the National Academy of Sciences 104(24): 9964-9969. (Scale bar: A, C, E, and G = 1 cm; B, D, F, and H = 500 mm [sic].) [They must mean microns.] 

The photomicrographs below illustrate the red oxide on some of the putative beads from Pigeon Cave. The reddish material adhering to some of the shells from the cave was chemically characterized as hematite. Hematite is also known as ochre. As I stated above, there appears to be little rhyme nor reason to these alterations.

From Bouzouggar, A., N. Barton, et al. (2007). "82,000-year-old shell beads from North Africa and implications for the origins of modern human behavior." Proceedings of the National Academy of Sciences 104(24): 9964-9969. (Scale bar: A, C, E, and G = 1 cm; B, D, F, and H = 500 mm [sic].) [They must mean microns.] 

At Blombos Cave the story is much the same. Total number is larger; different species; similar alterations. And I would make the same statement again: beyond the mere presence of fenestrations, their morphology and other modifications to these shells is hardly what I'd call patterned.

Henshilwood,C., d'Errico, F., et al. (2004). "Middle Stone Age Shell Beads from South Africa." Science 304:404. 

Fig. S1. (A) Juvenile (1), sub-adult (2-4) and adult (5) Nassarius kraussianus shells from the modern biocenosis of Duiwenhoks estuary (Cape Province, South Africa) showing gradual size increase and thickening of the lip (solid line). Blombos MSA beads (R) are all made on adult shells (scale bar, 1mm). (B) Perforation types observed on N. kraussianus shells from modern and archaeological collections. The shaded area indicates the variability in hole locations. Types 6 and 8 on the dorsal side, found respectively on 63% and 24% of MSA beads, are observed on 2% and 3% on natural shells (N=2587) from Duiwenhoks and Goukou River estuaries. Type 1 on the ventral side (P, Q), produced by the predator Natica tecta, is present on 60% of shells from modern thanatocoenoses (N=1836) and absent on MSA beads. (C) Perforations on the dorsal side of dead N. kraussianus from the Goukou estuary. (D-I) Macro- and SEM photos of use wear on two MSA N. kraussianus shells (E,H, scale bar, 1 mm; F,I, scale bar, 500μm). (J-O) Macro- and SEM photos of a modern (J-L) and LSA (M-O) N. kraussianus shell showing absence of wear facet on lip (K, scale bar, 1mm; L, scale bar, 200μm; N, scale bar, 500μm; O, scale bar, 200μm). (P-Q) Ventral aspect of a modern N. kraussianus with hole drilled by Natica tecta (Q, scale bar, 500 μm). (R) Perforations on two orange MSA N. kraussianus shell beads from Blombos Cave (scale bar, 1 mm). (S) Typical white N. kraussianus shell beads with large perforations from Blombos Cave LSA levels. Verbatim from Henshilwood,C., d'Errico, F., et al. (2004). "Middle Stone Age Shell Beads from South Africa." Science 304:404. 

So, at La Grotte des Pigeons and at Blombos Cave, archaeologists were compelled to interpret the origin and mode of deposition of these objects. I won't suggest that they were grasping at straws to give these finds importance from an evolutionary standpoint. I'll let you be the judge.

Both sets of authors use evidence from ethnography, natural history, or geology. For example, we know the shells don't occur naturally in these caves. Simple enough. True enough. But, how do we assess their assertion that the shells could have been purposefully transported to these sites only by humans or hominids?

Likewise the perforations. The authors are aware that these tiny mollusks have mollusk predators that are able to drill through the shell and feed on the animal within. In the image immediately above, in P and Q, you see a hole drilled by Natica tecta. The Blombos Cave excavators note that such artifacts of predation show up 60% of the time in the natural death assemblages of this species. They also point out that they see none of these perfectly drilled fenestra in their archaeological sample. What are we to make of that?

Of the facetting noted on several of the shells, both sets of authors claim that these must have arisen due to repeated contact between either bead and bead on a string of beads, or between some form of cordage or other material and the beads that were strung on.

Finally, the ochre. What can we make of the microscopic presence of hematite on some of the shells from La Grotte des Pigeons?

Let's take the transport agent to start with. Although the sea today splashes directly below Blombos Cave, that's not so for Pigeon Cave. The sea was also more or less close to Blombos cave at 80 ka. Not so for Pigeon Cave. If the same transport agent moved the shells from the seashore to the caves, it was capable of moving them several tens of kilometers. So. Was it humans/hominids, or was it something else? I'll admit that my efforts resemble more 'back-of-the-napkin' research than scholarly rigour, I was able to find at least one potential non-human transport agent capable of introducing such shells into the sediments at both caves---terns. Birds. The brown noddy [Anous stolidus] is known to collect such shells and incorporate them in their nests. The noddy is a tern. Terns occur throughout the world, and the brown noddy is native to South Africa and Morocco, as well as almost everywhere else in the world. How could one possibly rule out the possibility that these birds, or other occasional collectors, introduced these items to the archaeological context?

The authors would say that the nature and placement of the holes suggests human agency and the polish and traces of hematite support that conclusion. What about those holes? For starters, another quick trip to the internets tells me that at least two groups of animals are capable of boring through mollusk shell and that it is a commonplace the world over. The one below is riddled with tiny perforations caused by a species of sponge. Yup. Sponge. In addition, as you'll have learned from the Henshilwood et al. paper, many species of predatory mollusks bore through other mollusk's shells to get at the living creature within. You'll be thinking "Yes, but the archaeological shells from Pigeon Cave and Blombos are not as beautifully regular as those left by sponges, or, for that matter by Natica tecta." True enough. They appear to be ragged, not drilled. More on that after the pretty picture.

Far be it for me to accuse palaeo people of having a courser aesthetic sense than I, but the majority of the perforations claimed to be the result of human/hominid activity look as if they've been punched out by pressure applied either from the outside or from the inside. Who's to say that those rough perforations weren't the result of inadvertent pressure [underfoot, for example, or in pounding surf]? It's logical to assume that pressure placed on a part of the shell where a sea creature's boring had previously weakened that portion of the shell would be capable of punching out a larger hole. Damned difficult to choose. Don't you think? Natural? Or not?

I realize that I'm merely casting doubt on the conclusions of the Blombos and Pigeon Cave archaeologists. Unlike the law, I can't employ the 'beyond a reasonable doubt' criterion when I'm interpreting an archaeological site. But let's face it. These claims are at best tentative, and would require, I believe, extensive experimental, ethnoarchaeological, and ethological examination before we accept them as fact.

Ooops. almost forgot. The polish. I think the presence of polish on just a few specimens should rule out any notion that they'd been strung like beads on a necklace. At a minimum the authors would want to examine ethnographic strings of similar or identical shells before claiming that the rare facets they find are due to rubbing of one shell against another on a string. Moreover, if such shells are susceptible to wear when suspended on a string, they can just as easily get worn on a beach that comprises numerous other such shells. After all, they all originated at the water's edge. And, although the authors acknowledge that certain kinds of damage might be possible at the beach, their investigations of present-day death assemblages of these animals reveals to them that the modifications to the archaeological specimens are unlike those occurring on present-day strands. To that I would only say that time passes and circumstances change. What effect would over-harvesting of sponges have on the frequency of perforations on modern day mollusk death assemblages in Morocco? Or pollution in the Mediterranean? Or different substrates at different times at Blombos? I think you get my point.

And finally the ochre. Again, if ochre can occur in a cave at a distance from the ocean, it can occur at the sea-side.

For all of the reasons I've given here, it's clear to me that the claims from La Grotte des Pigeons and Blombos Cave should be viewed with extreme caution, if only because they're extreme claims for modern human behaviour at extreme time depth.

I hope you enjoyed this little road trip to the seaside. See you *cough* next time.

SA announces new posts on the Subversive Archaeologist's facebook page (mirrored on Rob Gargett's news feed), on Robert H. Gargett's Academia.edu page, Rob Gargett's twitter account, and his Google+ page. A few of you have already signed up to receive email when I post. Others have subscribed to the blog's RSS feeds. You can also become a 'member' of the blog through Google Friend Connect. Thank you for your continued patronage. You're the reason I do this.

Stay-At-Home Modern Humans Stranded in South Africa for 50,000 Years. What's Up With That?

Pinnacle Point Caves (PPC) in relation to Klasies River Mouth (KRM)  cave and Blombos Cave (BBC), three of the best known southern African localities with modern human accoutrements as early as 70 ka to 100+ ka. 

In the midst of the recent science-media frenzy about bows and arrows at Pinnacle Point, I thought I'd best expand on my reasons for doubting, en masse, the 70 ka to 90 ka [and beyond] dates for what appear to be unequivocally modern behaviours in a growing number of southern African archaeological sites. I've been working on this for several days, and the more I look, the more I see evidence of a real problem with optically stimulated luminescence (OSL), the technique used at all of these sites to estimate the age of the deposits. As I've mentioned previously [e.g. here], the primary assumption underpinning OSL age estimates is this: each individual quartz grain removed from the archaeological deposits MUST have been exposed to sunlight for long enough to have released any free electrons trapped in its crystal lattice. Given that all of these early dates derive from cave deposits, the primary issue for the OSL specialist is eliminating any quartz grains that have not been exposed in the manner just described. It's one thing to recognize this pesky problem; it's quite another to overcome it. And that, I think, is where OSL falls down. Whether or not a given age estimate is accurate becomes a real guessing game, as we'll see in a moment.  

Three representations of the stratification at Pinnacle Point PP5-6. Source: Brown et al. 2012

As you can see above, in the recently published profile from Pinnacle Point site PP5-6, the OSL picture is evidently not straightforward. The date of 53 ka highest up in the profile is superposed on a date of 60 ka. That is consistent with the stratigraphic relationship between the sediments that yielded the dates. However, the next lowest in the column is, again, 53 ka. Then come 56 ka, 52 ka, 50 ka, followed by 52 ka, 60 ka, 65 ka, 65 ka, 69 ka, then 66 ka, 66 ka, 64 ka, 72 ka, 71 ka,74 ka, 69 ka, 77 ka, 73 ka, 71 ka, 69 ka, 69 ka, and lastly 78 ka. Clearly this is not an 'internally consistent' temporal sequence. These age estimates are roughly sorted youngest to oldest, but given the lack of coherence, how is one to assess, much less find acceptable, either the youngest or the oldest dates?
     For example, does one conclude that the real age of this stratigraphic column ranges from 53 ka to 78 ka? How about 53 ka to 69 ka? Or 60 ka to 69 ka? Moreover, with such variance from what one should expect of the age succession in a sedimentary column that has maintained temporal integrity, how do we know that any of these estimates are accurate? All this points to, in my barely savvy opinion, serious difficulty with the technique, most likely due to the intractable problem inherent in the primary assumption of the technique. And, if the very latest results of the best of the best practitioners yields such shaky outcomes, what does it say about all of the other South African site chronologies? Here, then, lies the very real possibility that the OSL technique is single-handedly creating a false picture of modern human origins throughout the southern region.

Stratigraphic profile of La Quina, the type site of the Quina Mousterian. All of the objects illustrated are éboulis, clasts of all sizes and shapes, which have spalled off the bedrock and been incorporated along with the archaeological materials.

     In this I'm reminded of what's now a discredited relative dating technique that European archaeologists depended on for a good long time. I refer to the technique of éboulis analysis that underpinned [especially] the French Palaeolithic chronology for much of the late twentieth century. Briefly, the analyst qualitatively assessed the angularity of small pieces of bedrock breakdown and assigned an age estimate that coincided with what was then considered to be a four-stage European glacial cycle--Gunz, Mindel, Riss, Würm. The most angular pieces represented cold or dry or both, and the most rounded pieces warm and wet or both. It seems a bit naïve by today's standards. However, it was gospel for decades. One day it all came crashing down. After that, everyone picked up the pieces and moved on. Thus far archaeologists of the Middle Pleistocene haven't resorted to OSL for dating. However, they do use thermoluminescence, which depends not on individual quartz grains, but rather on pieces of flint that have been subjected to high heat. There's certainly less doubt about whether or not a piece of flint has been heated than that a tiny grain of quartz was ever exposed to sunlight. 
     In making this comparison between éboulis analysis and the OSL dates from southern Africa, I don't mean to suggest that the two techniques are equal. Rather, I make the comparison because, in the past, an error-prone technique was used to provide the chronological framework for an entire sub-continent. That's the direct comparison to be made. The southern African cave sites are universally dated using OSL or other equally shaky [to my mind] techniques.

The crux of the southern African OSL dates. In relation to the earliest evidence of modern human behaviour everywhere outside of Africa, and the speed at which Eurasia was 'colonized', the southern African dates stand out like a sore thumb. 

     Having made the comparison between éboulis and OSL, I can now introduce what I believe to be several data sets suggesting that the southern African cave age estimates are anomalous when compared to the rest of the world. For, when it comes to the early dates for modern human behaviour in southern Africa there is an absence of parsimony. Parsimony, if you'll recall, is the central point of Occam's Razor, the idea that, when deciding which competing hypothesis is the more correct, the one that makes the fewest assumptions should be selected. Occam's Razor is by no means a law of nature. It merely crystalizes what empirical scientists have know for some time--the more auxiliary hypotheses you need to support your main conclusion, the less likely it is to be the right one. In the case of the early dates for modern human behaviour in South Africa, there is much that flies in the face of Occam's Razor.
     Take, for example, the overwhelming evidence for the almost instantaneous spread of behaviorally modern humans to the furthest reaches of Asia, which happened about 45 ka. Australia was colonized at this time. So were some of the the furthest reaches of Europe--England, for example, at 44 ka. The best explanation for the absence of modern human archaeological assemblages prior to this time is that there simply were no modern humans outside of Africa prior to that time. The question thus becomes, if humans could expand the length and breadth of Asia in an eye-blink of geological time, why didn't it happen earlier, when modern human behaviour is claimed to have had its start 70 ka to 90 ka ago in southern Africa?
     To explain the apparent lag between the emergence of modern human behaviour in southern Africa and that which inhered everywhere else in the world post 45 ka, the various archaeological teams must invoke untested, and indeed untestable hypotheses as to the reason for those moderns staying put for anywhere between 164 ka and [taking the youngest date from PP5-6] 53 ka. One hypothesis has that population bottlenecks occurred due to environmental circumstances [as yet unrecorded], after which cultural achievements of the modern populations were stultified, only to reappear some 30,000 years later, at which time the archaeology records the rapid spread of modern humans throughout what's been called the Old World.
     In July of this year we saw a report from Border Cave of a modern human archaeological assemblage that dates to about 44 ka, a result in accord with the beginning of modern human behaviour in Eurasia. But, because they found no earlier deposits, the authors find it necessary to replay the tape of the 'early' teams, that the Border Cave moderns must represent the hypothesized re-fluorescence of modern humans preparatory to world domination. Once again. I think such intellectual reaching is unnecessary, at best, and obfuscatory at worst. I do not mean to suggest that the obfuscation is deliberate. The archaeologists who rely on OSL age estimates have nothing else to go on. Or, at least, they have no reason to think that their information is in error. After all, the best of the best are working at Pinnacle Point. Who could doubt their veracity? Who would, much less?

Modern human artifacts from Border Cave at 44 ka. Touted to be the earliest evidence for the same culture that existed ethnographically in the vicinity of Border Cave. [Take that with an anthropological grain of sand.] Source: D'Errico et al. 2012.

     I know. I know! We're dealing with Culture. And culture theory says that the early moderns in southern Africa might just have decided that they wanted to stay put and not colonize the known universe. If you think about it, how would they have known they were choosing not to colonize the rest of the world, unless they knew there was a 'rest of the world' to colonize. And, if they did know that there was a 'rest of the world,' why is there no evidence for those modern humans until about 45 ka everywhere but southern Africa. If, and I have no doubt, the early moderns in South Africa were anything like modern moderns, or a great many ethnographic moderns, they couldn't have resisted seeing what was beyond the next ridge. Yet it would seem that they 'chose' to stay put, enjoying the fruits of their modernity all by themselves, thus endangering their persistence by allowing themselves to be buffeted by the whims of the environment. That doesn't sound like modern me to me. [N.B.: I'm only being half facetious.]
     I supposes one could suggest that there were natural barriers to their spread out of southern Africa. But, there are no natural barriers to movement in coastal Africa, nor would there have been when the sea levels were depressed during the glacial stages of the Pleistocene. If they could process pigments and decorate ochre, make beads, and arrows, they were equally capable of making boats to cross the broad reaches of Africa's rivers where they meet the sea. After all, the colonizers of Australia crossed at least 90 km of open ocean at around 45 ka. Surely the precocious moderns in south Africa at 70 ka to 90 ka could have done the same. So, the big question is why did they stay put in southern Africa? Huh? Riddle me that, Batman. Apparently the Wall Street Journal has a suggestion.

     So, you see, reason alone tells me that there's much wrong about the early dates from southern Africa. And my instinct tells me that the OSL dates must be wrong. As I've said to you before, will some kindly physicist not come forward to examine the mathematical filmflammery that accompanies OSL age estimates, and thereby save me from this empirical purgatory??? The flimflammery in question.

Please??? I'm beggin' ya.


........ 

Thanks for dropping by! If you like what you see, follow me on Google Friend Connect or Twitter, friend me on Facebook, check out my publications at Academia, or connect on Linkedin. You can also subscribe to receive new posts by email or RSS [scroll to the top and look on the left]. And please don't forget. Oh, and you can always put me on your blogroll! By the way, I get a small commission for anything you purchase from Amazon.com if you go there by clicking through from this site.