There Ya Go Again! Part One–The Putative 500,000-Year-Old Hafted Spear Points From Kathu Pan 1 VS Reality: What’s the DIF?

It’s now officially an international media feeding frenzy.
CBC.ca –  Stone-tipped spears used by human hunters much earlier in time
National Geographic –  Stone Spear Tips Surprisingly Old—”Like Finding iPods in Ancient Rome”
Science News – Oldest examples of hunting weapon uncovered in South Africa
New Scientist – First stone-tipped spear thrown earlier than thought
Daily Mail – The birth of weapons: Researchers discover man began hunting with stone …
Voice of America – Archeologists Identify Oldest Spear Tips
Scientific American – Human Ancestors Made Deadly Stone-Tipped Spears 500000 Years Ago
Christain Science Monitor – Gone spear hunting: Ancestors used stone spear tips 500000 years ago
Times of India – Man began spear-hunting 200000 years earlier than believed
U.P.I. – Stone-tipped weapons older than thought
Vancouver Sun – Scientists spear animals to make point [my home-town newspaper. *sniff*]
China Post – Early human beings used stone spear points at least 500000 years ago: experts
Windsor Star – Spear study gets to the point
The Independent – Prehistoric arms race started earlier than previously thought
The Daily Telegraph – Man hunted with spears half a million years ago
News 24 – SA spear tip made by earlier ancestor
The Seattle Times – Prehistoric man became deadlier earlier than previously thought
HuffingtonPost.com – Stone Spear Tips Suggest Weapons Were Developed Much Earlier Than Previously Thought
Science News – Archaeologists Identify Oldest Spear Points: Used in Hunting Half-Million Years Ago
Business Week (!) – Prehistoric Man Was Deadlier Earlier Than Once Thought
Salon.com – Study: Stone spear tip made by earlier ancestor

From Wilkins et al. 2012
This is beginning to feel like one of those squirmily popular horror series like Hallowe’en. You know? Freddie of the long fingernail knives. A tip o’ the hat to Physorg for pointing us in the the direction of the November 16 print edition of Science, where one finds this:

Evidence for Early Hafted Hunting Technology,’ Jayne Wilkins, Benjamin J. Schoville, Kyle S. Brown, and Michael Chazan. Science 338, 942-946, 2012; DOI: 10.1126/science.1227608

Yes, sports fans, the same Jayne Wilkins who gave the Subversive Archaeologist fits a while back [here, herehere, and here] with her claim of a ‘blade industry’ at Kathu Pan 1 in South Africa. It’s also the same Kyle Brown who was the lead author on the Pinnacle Point microliths and ancient heat-treating of lithic raw material. [South African archaeology is obviously a closely knit community.] These latest claims from Kathu Pan 1 are made for stone tools recovered from the same site as the earlier ‘blade industry,’ an infilled sinkhole, or doline. This time around, some pointy artifacts with step fractures on the distal (pointy) end and huge proximal (butt) ends are claimed to be evidence for, as the title would suggest, composite tools. And, once again, the claim is made for a VERY early date, and I quote ‘~500 ka, coeval with H. heidelbergensis [or H. antercessor, if you prefer].’
     I hope you’ll forgive what will undoubtedly end up being a very long blurt this time around. It’s dead necessary. If your Subversive Archaeologist is going to spot shortcomings that undermine the arguments made in a paper like this, he’s going to have to spend way more time, and marshall ten times the breadth of expertise than did the referees. And, remember, these are Science referees. There’re no flies on them [at least in the opinion of the Science editor(s)]! So, this is not a task that lends itself to brevity.

Vicinity of Kathu Pan 1 (from Wilkins et al. 2012). 

The authors are well aware that such claims cannot be made simply by bold assertion. In today’s case, the extraordinary conclusions are accompanied by some truly impressive morphological and metrical analysis, and actualistic experiment with a purpose-built crossbow, some hafted spear stand-ins, and a deceased wild springbok [small ruminant artiodactyl]. Unfortunately for the clever archaeologists, the basis of their claim rests not in their analysis, but rather in their arguments. It’s the same old story that the Subversive Archaeologist has told before. Funny thing is…few, it would seem, are listening. Maugre their ultra-scientistic approach.

Notice the veritable totem pole used to ‘analogize’ a throwing weapon. The ‘business end’ is a flake hafted to a 1-1/8″ (2.8575 cm) clothes-closet dowel, inserted into a piece of 1″ galvanized pipe joined to a piece of 1-1/4″ pipe. Not that it makes any difference to their conclusions, but this is less analogous to a javelin than it is a small tree-trunk! (From Wilkins et al. 2012)

I’ll save my mirthful response to the use of a springbok for such an experiment for a quiet moment alone in my mother’s basement. But, seriously, unless the thing were lying dead on the pan, can you imagine H. heidelbergensis [or you or me, for that matter] sneaking up on a nimble beast like this one and driving home such a weapon? They might as well have chosen a dead elephant [but that might not have impressed the authorities].
     This article has two major shortcomings, as I see it. The first is in convincing a sceptical archaeologist that the dates are from contexts in which the association with artifacts is well-warranted. Their second mistake comes in the background knowledge that they draw on in making their claim that the only way their bits of stone could have been modified in the way they were because they were hafted to a stick and poked into an animal. I’ll deal with the dates in this part; the stones themselves will have to wait.
     Wilkins et al.’s troubles have their origin in the Kathu Pan 1 doline’s depositional history, and in their evidently weak expertise in interpreting the depositional agents of various sized clasts. There is no indication that the authors are even aware of the potential for mixing in a sediment trap like the one in which they’re excavating, much less the action of the occasional spring that came into being during the site’s depositional history. The authors appear to treat each excavated artifact and sand grain chosen for OSL [yup, again] as if they had been, for all time, temporally associated. How else could you explain their naïve acceptance of the OSL age estimates?

Filled circles are OSL age estimates; filled triangle is U-series/ESR age estimate. From Wilkins et al. 2012.

To get the critical ball rolling, check out the schematic excavation profile above. The authors describe the sediments thusly:

From top to bottom, stratum 1 [not shown] is characterized by 1.5 -2 m of interdigitating calcified sand and organic peats. Stratum 2 [not shown] is characterized by well-sorted aeolian sand that becomes increasingly calcified toward the top. Artifacts in strata 1 and 2 are very sparse, but Beaumont tentatively suggests a ceramic LSA or Iron Age designation for stratum 1 and potential Robberg affiliation for stratum 2.

In other words, too young to be of interest to the authors.

Stratum 3 is … gravel with sub- angular to sub-rounded pebbles in a greyish sand matrix. 

Stratum 3 is dated at 291 ka.

Stratum 4 consists of two substrata, 4a and 4b, which are distinguished from each other based mainly on the lithic and faunal assemblages, though Beaumont … also reported a thin pebble lens that divided the substrata. The artifacts were recovered from a yellowish sand matrix.

Stratum 4 sounds like well sorted sand. Most probably wind-borne. In fact, I’d bet my lower central left incisor that these are fine to medium-sized sand grains, typical of aeolian transport and deposit. [I should tell you that I can make good on this bet. The orthodontist that the tooth be extracted before they put braces on my pearly whites, and I still have it under my pillow, waiting for the tooth fairy.] Moreover, I’d bet that the period during which the sands were deposited was an extraordinarily dry and windy one. That’s because there is evidence of what could only have been scouring out by wind in the non-conformity between 4 and 3. Stratum 3’s coarser, unsorted sediments bespeak a later, wetter period when surface water flow was transporting a range of small-sized sedimentary grains into the doline.
     But, here’s the kicker. The really old dates are from an intrusive geomorphic feature, one that would undoubtedly have brought older fine sediments up to the level of the excavations, and described by the authors as follows.

In the process of cleaning the section, two well-defined vertically-oriented spring vents in stratum 4a were revealed and described as the ‘Upper Vent’ and ‘Lower Vent’. The Upper Vent, which is in the uppermost levels of stratum 4a and truncated at the top by Stratum 3, is densely packed with lithic artifacts and fauna, and the area outside the vent contains few, if any, artifacts. [The 464 ka sample] was recovered from sediments in direct association with the concentrated lithic artifacts and faunal remains within the vent. 

What does this tell us? And how does this relate to the veracity of Wilkins et al.’s claims? Plenty.

A brief accounting of karst features.

To begin with, think of how a big hole in the ground [the doline] came to be, and then filled up. There’s a clue in the name of the site, Kathu Pan. A ‘pan’ is what would be called a playa in North America. It’s a broad, nearly horizontal valley bottom that seasonally collects water from precipitation, hosts a shallow lake for a time, then dries up and awaits the next wet season.
     For a doline to form, the pan must be underlain by lithified fine sediments, such as limestone, siltstone or mudstone. These are the kinds of geological strata that are susceptible to karst processes–cave formation, among others. The illustration above is a succinct depiction of the kinds of geomorphic features that can occur in areas of karst. Those below are real-life dolines.

Guatemala City, Guatemala, 2007. A ‘swallow hole.’ It will need to be bridged. Filling it in is out of the question.
Florida. This is more of a ‘sink’ than a sinkhole. But just as devastating. 

A doline [or sinkhole] is the surficial expression of a subsurface void, or cave, the ‘roof’ of which became too thin near the surface and collapsed. Sinkholes are common in the middle of the US, and anywhere that the bedrock can be dissolved by surface water percolating downward. [The Santa Cruz campus of the University of California, where I’ve worked for the past 8 or so years, is built on mudstone that hosts a cave system, and in many places there are sinkholes at the surface.]
     So, once formed, the Kathu Pan 1 doline and its associated solution features began to fill in with any sediments that were transported by wind or water to its gaping maw. According to the authors, the pan in question is still somewhat depressed relative to the surrounding surface, and is thus still filling in. [It’s likely that, as the underlying bedrock evolves, voids get larger, leading to settling of the fill material. Thus, a feature like the Kathu Pan 1 doline will always be a dynamic sedimentary environment, and might never be, truly, filled in.]
     The basal geology of Kathu Pan 1 would be of no intrinsic value to this discussion of the OSL age estimates were it not that Wilkins et al. observed ‘spring vents’ in their site. Your subversive imagination should, upon hearing such a statement, see a red flag. A spring vent is an opening in the surrounding sediments that concentrates ground-water discharge to the Earth’s surface. In other words, it’s like a do-it-yourself geological pressure release valve. In the case of Kathu Pan 1 these spring vents could only have occurred at a time when the underlying karst system was chock full of water under pressure from the local water table. Some of that pressure was relieved through a ‘vent’ that formed in the unconsolidated sediments of the doline’s fill. Think of it. Everything in the vent would be turned into a slurry as the water passed through it toward the surface. As in any transport process involving moving water [even water moving vertically through sediment] the smaller and lighter clasts would be more mobile, and the heavier pieces would tend to form a lag deposit. [Exactly fitting the authors’ description: ‘concentrated lithic artifacts and faunal remains.’] In a vertical column that contains fines like sand and larger pieces like stone artifacts, the heavier artifacts would literally sink as the smaller grains were randomly removed from beneath them–just the way your feet sink when you’re standing on a beach and a wave washes ashore and then recedes.

     It simply can’t be argued away or ignored. The authors’ concentration of artifacts and faunal remains is almost certainly the result of the these processes I’ve described. I’m not makin’ this stuff up! Karl Butzer, one of the lions of geoarchaeology in the early days, thought that a spring vent was an important enough disturber of archaeological traces that he put a wonderful illustration in Archaeology as Human Ecology, shown below.
     So, quite apart from the claims of hafted stone artifacts, the authors have a big problem with their dates. If it isn’t already patently obvious [and there’s no reason it should be] the dates from Stratum 3 and the spring vent are suspect. We have no way of knowing the stratigraphic origin of the coarser sands and pebbles in Stratum 3. All we know is that they came from somewhere else further up the fluvial train. They could have been eroded from the older sediments in which the doline formed, or any deposit created since then. How to know? [More OSL mathematical flim-flammery, I suspect.]
     In a different way, but with the same result, the fines selected for OSL dating from the spring vent were undoubtedly a porridge of old and young clasts. Since the water was pushing upward through older sediments, we can have NO WAY OF KNOWING where they came from or indeed how and if they are related temporally to the artifacts that almost certainly derived from a younger level or levels. The spring vent could be as recent at Holocene in age, given the other uncertainties raised by the depositional circumstances in Stratum 3. Bottom line: we could never know the true age of any artifact in the spring vent that passed through Stratum 4 (a and b).

There is no question that a spring vent, such as that occurring at Kathu Pan 1, would wreak havoc with the sediments in its path, and those nearby.

I’ve raised a lot of questions in this piece. So, to summarize, the OSL dates from Kathu Pan 1 cannot be relied upon. Because of the depositional environments–a sinkhole and a spring vent–there are too many unknowns when it comes to the association between the artifacts that interest the authors and the sediments selected for dating. Too. Many.

     I’m truly sorry if this comes across as a harangue. I get so frustrated by the way people working in my business are so concerned to get a spectacular result that they ignore very basic principles of stratigraphy and geomorphology, to say nothing of informal logic. The present authors are no exception [as we’ve seen before].
     Watch this space for a discourse upon the stones, coming soon.

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. 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s